Cytokine Protocols - MArc Ley

METHODS
IN
MOLECULAR BIOLOGY
Series Editor
John M. Walker
School of Life Sciences
University of Hertfordshire
Hatfield, Hertfordshire, AL10 9AB, UK
For further volumes:

http://www.springer.com/series/7651
Cytokine Protocols
Edited by
Marc De Ley
Katholieke Universiteit Leuven, Heverlee, Belgium
Editor
Marc De Ley
Katholieke Universiteit Leuven
Afd. Biochemie
Celestijnenlaan 200 G
3001 Leuven
Belgium
marc.deley@chem.kuleuven.be
ISSN 1064-3745
e-ISSN 1940-6029
ISBN 978-1-61779-438-4
e-ISBN 978-1-61779-439-1
DOI 10.1007/978-1-61779-439-1
Springer New York Dordrecht Heidelberg London
Library of Congress Control Number: 2011940833
Springer Science+Business Media, LLC 2012
All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the
publisher (Humana Press, c/o Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA),
except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information
storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or
hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified
as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights.
Printed on acid-free paper
Humana Press is part of Springer Science+Business Media (www.springer.com)
Preface
Seven years have passed since the first volume of Cytokine Protocols was published in the
series of Methods in Molecular Biology (Volume 249, 2004) of Humana Press. Since
then, not only the number of known/characterized cytokines has drastically increased (e.g.,
interleukins up to IL-35) but also assays for gene expression have become more sensitive
and sophisticated, allowing the simultaneous processing of higher numbers and/or smaller
samples. In recent years we have witnessed the far-advanced miniaturization of microanalytical methods as well as the extensive development of bioinformatics and nanotechnology.
Together these allow performing methods such as genomics, transcriptomics, and proteomics. At the same time a substantial reduction in sample size was achieved, allowing
accurate determinations that were previously impossible. Single-cell based assays are
expected to further extend this broad range of assays.
The first chapter written by my colleague professor A. Billiau is not only of historical
importance but also brings a message of general importance to researchers using bioassays
in general. Careful observation and interpretation of results obtained in two different
(biological) assays with respect to possible differences may reveal the presence of other
hitherto unknown cytokines to be discovered and further characterized.
The next three chapters deal with the quantification and characterization of cytokinerelated RNAs. These range from the cytokine mRNAs themselves over cytokine-induced
genes until miRNAs. Real-time quantitative PCR (RT-qPCR), now widely established as a
standard molecular biological technique, yields accurate determinations of single cytokine
mRNA transcript levels (Chapter 2). Simultaneous measurement of gene expression profiles after cytokine stimulation is made possible through application of DNA microarray
techniques (Chapter 3). The eventual level of mature active mRNA depends on multiple
regulatory factors and processes, among which miRNAs. Their accurate quantitative determination (as well as that of their precursors) can also be executed by RT-qPCR (Chapter 4).
The next seven chapters deal with the posttranscriptional modifications of RNA, taking
place either naturally or artificially. One of the most decisive factors in determining cytokine
levels and the response to it are mRNA levels, themselves being regulated by two opposite
mechanisms: generation and decay, in turn regulated by cis-elements as well as trans-acting
proteins. Both their characterization and evaluation yields further insight in the signal
transduction processes (Chapter 5). One of the well-known mechanisms acts through the
interaction of proteins with AU-rich elements in the 3UTR of mRNAs, the involvement of
which can be demonstrated using a cell-based GFP assay (Chapter 6). Although the highly
selective and efficient silencing of genes by siRNAs is known already for a long time, the
delivery of these siRNAs to some kinds of cells restricts its broader application. A neat way
to overcome this obstacle is through their inclusion in (integrin) targeted stabilized nanoparticles (Chapter 7). Another proven method for gene silencing is through the application
of carefully designed and validated hammerhead ribozymes. These can either be introduced
in the cell as chemically modified ribozymes (in order to increase their half-life) or be constitutively generated in situ by appropriate plasmids (Chapter 8). A well-known and often
vi
Preface
undesirable side effect of RNAi methodology is the induction of interferon response, either
by the production of the cytokine itself or by the induction of interferon-related gene transcription. Hence, it is often difficult to distinguish between the pursued RNAi effect and
the confusing interferon effects (Chapters 9 and 10). RNAi technology allows very specific
targeting to a particular gene transcript and hence to a specific member in a signal transduction pathway. This very powerful approach is, however, often hampered by difficulties
encountered at the introduction of the foreign DNA in the recipient cell (hard-to-transfect
cells, e.g., primary cells) and by its possible toxicity. Therefore, different protocols and
reagents should be carefully compared (Chapter 11).
The last three chapters are devoted to observations at the protein level. Following the
identification of a novel cytokine biological activity, the next big challenge is the isolation,
purification, and characterization of its first contact with the cell, i.e., its membrane receptor. Ligand affinity chromatography is the method of choice, allowing in most cases the
isolation of sufficient amounts of intact receptor for partial sequence determination
followed by full sequence prediction from data banks. Moreover, this method may also lead
to the discovery of unexpected, unpredicted (non-receptor), interacting proteins
(Chapter 12). Accurate and sensitive detection of cytokine levels is of prime importance in
the evaluation of their biological activity, both in situ (intracellular) and in vitro (solution)
methods are needed. Application of fluorescently labeled monoclonal antibodies in combination with flow cytometry on permeabilized cells allows sensitive detection even in individual cells (Chapter 13). As already explained in the first chapter, sensitive and specific
detection of the biological activity of cytokines is of utmost importance. It is well known
that each cytokine is quantified most specifically, accurately, and with the lowest detection
limit on a different cell type, thus obliging researchers that work with different cytokines to
maintain a whole series of cultures of various cells, each with their own detection system).
This problem can be partly circumvented by constructing cell lines with chimeric receptors,
the extracellular part of them being specific for each cytokine, the intracellular part being
the same for all and thus requiring only one kind of signal detection (Chapter 14).
Heverlee, Belgium
Marc De Ley
Contents
Preface. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Contributors. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
v
ix
1 A Tale of Two Interferon Bioassays: How Frustration with Discrepant Results

from Slightly Dissimilar Methods Can Engender Discovery. . . . . . . . . . . . . . . . . . .
1
Alfons Billiau
2 The Use of Real-Time Quantitative PCR for the Analysis of Cytokine
mRNA Levels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7
Maria Forlenza, Thomas Kaiser, Huub F.J. Savelkoul,
and Geert F. Wiegertjes
3 Interleukin-27 Induces Interferon-Inducible Genes: Analysis of Gene
Expression Profiles Using Affymetrix Microarray and DAVID . . . . . . . . . . . . . . . . .
25
Tomozumi Imamichi, Jun Yang, Da Wei Huang, Brad Sherman,
and Richard A. Lempicki
4 Quantitative Analysis of miRNA Expression in Epithelial Cells and Tissues . . . . . . .
55
Markus Bitzer, Wenjun Ju, Xiaohong Jing, and Jiri Zavadil
5 Evaluating Posttranscriptional Regulation of Cytokine Genes . . . . . . . . . . . . . . . . .
71
Bernd Rattenbacher and Paul R. Bohjanen
6 Cloning of Cytokine 3 Untranslated Regions and Posttranscriptional
Assessment Using Cell-Based GFP Assay. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
91
Latifa Al-Haj and Khalid S.A. Khabar
7 Integrin-Targeted Stabilized Nanoparticles for an Efficient Delivery
of siRNAs In Vitro and In Vivo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
Charudharshini Srinivasan, Dan Peer, and Motomu Shimaoka
8 Hammerhead Ribozyme-Mediated Knockdown of mRNA for Fibrotic
Growth Factors: Transforming Growth Factor-Beta 1 and Connective
Tissue Growth Factor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Paulette M. Robinson, Timothy D. Blalock, Rong Yuan,
Alfred S. Lewin, and Gregory S. Schultz
9 Control of the Interferon Response in RNAi Experiments. . . . . . . . . . . . . . . . . . . . 133
Jana Nejepinska, Matyas Flemr, and Petr Svoboda
vii
viii
Contents
10 shRNA-Induced Interferon-Stimulated Gene Analysis. . . . . . . . . . . . . . . . . . . . . . . 163

Nria Morral and Scott R. Witting
11 Use of RNA Interference to Investigate Cytokine Signal Transduction
in Pancreatic Beta Cells. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
Fabrice Moore, Daniel A. Cunha, Hindrik Mulder, and Decio L. Eizirik
12 Ligand Affinity Chromatography, an Indispensable Method
for the Purification of Soluble Cytokine Receptors and Binding Proteins . . . . . . . . . 195
Daniela Novick and Menachem Rubinstein
13 In Vitro Stimulation and Detection of IFNa Production in Human
Plasmacytoid Dendritic Cells. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
William C. Adams and Karin Lor
14 A Sensitive and Versatile Cytokine Bioassay Based on Type I Interferon
Signaling in 2fTGH Cells . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
Lennart Zabeau, Jos Van der Heyden, and Jan Tavernier
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
Contributors
WILLIAM C. ADAMS Department of Medicine, Center for Infectious Medicine,
Karolinska University Hospital Huddinge, Stockholm, Sweden
LATIFA AL-HAJ Program in Biomolecular Research, King Faisal Specialist
Hospital and Research Center, Riyadh 11211, Saudi Arabia
ALFONS BILLIAU Rega Institute for Medical Research, University of Leuven
(Katholieke Universiteit Leuven), Leuven, Belgium
MARKUS BITZER Internal Medicine, Nephrology, Michigan Diabetes Research and
Training Center, University of Michigan, Ann Arbor, MI 48109, USA
TIMOTHY D. BLALOCK Department of Obstetrics and Gynecology,
College of Medicine, Institute for Wound Research, University of Florida,
Gainesville, FL 32610-0294, USA
PAUL R. BOHJANEN Department of Microbiology, Center for Infectious Diseases and
Microbiology Translational Research, University of Minnesota,
Minneapolis, MN 55455, USA
DANIEL A. CUNHA Laboratory of Experimental Medicine, Universit Libre
de Bruxelles, Brussels BE-1070, Belgium
DECIO L. EIZIRIK Laboratory of Experimental Medicine, Universit Libre
de Bruxelles, Brussels BE-1070, Belgium
MATYAS FLEMR Institute of Molecular Genetics, Academy of Sciences
of the Czech Republic, 142 20 Prague 4, Czech Republic
MARIA FORLENZA Cell Biology and Immunology group, Department of Animal Sciences, Wageningen University, Wageningen PG 6709, Netherlands
DA WEI HUANG Laboratory of Immunopathogenesis and Bioinformatics,
CSP, ADD, SAIC-Frederick, Inc., NCI-Frederick, Frederick, MD 21702, USA
TOMOZUMI IMAMICHI Laboratory of Human Retrovirology, Clinical Services
Programs (CSP), Applied Developmental Directorate (ADD),
Science Applications International Corporation (SAIC)-Frederick, Inc.,
National Cancer Institute (NCI)-Frederick, Frederick, MD 21702, USA
XIAOHONG JING Computational Biology Center, Memorial Sloan-Kettering
Cancer Center, New York, NY 10021, USA
WENJUN JU Internal Medicine, Nephrology, Center for Computational Medicine and
Bioinformatics, University of Michigan, Ann Arbor, MI 48109, USA
THOMAS KAISER Cell Biology and Immunology group, Department of Animal Sciences,
Wageningen University, Wageningen, PG 6709, Netherlands
KHALID S.A. KHABAR Program in Biomolecular Research, King Faisal
Specialist Hospital and Research Center, Riyadh 11211, Saudi Arabia
RICHARD A. LEMPICKI Laboratory of Immunopathogenesis and Bioinformatics,
ALFRED S. LEWIN Department of Molecular Genetics and Microbiology,
University of Florida, Gainesville, FL, USA
ix
Contributors
KARIN LOR Department of Medicine, Center for Infectious Medicine,

Karolinska University Hospital Huddinge, Stockholm, Sweden
FABRICE MOORE Laboratory of Experimental Medicine, Universit Libre
de Bruxelles, Brussels, BE 1070, Belgium
NRIA MORRAL Department of Medical and Molecular Genetics,
and Center for Diabetes Research, Indiana University School of Medicine,
Indianapolis, IN 46202, USA
HINDRIK MULDER Unit of Molecular Metabolism, Department of Clinical
Sciences in Malm, Lund University Diabetes Center, Clinical Research
Center 91:12, Malm, SE 205 02, Sweden
JANA NEJEPINSKA Institute of Molecular Genetics, Academy of Sciences
DANIELA NOVICK Department of Molecular Genetics, The Weizmann Institute
of Science, Rehovot 76100, Israel
DAN PEER Laboratory of Nanomedicine, Department of Cell Research
& Immunology, George S. Wise Faculty of Life Sciences, and the Center
for Nanoscience and Nanotechnology, Tel Aviv University, Tel Aviv 69978, Israel
BERND RATTENBACHER Department of Microbiology, Center for Infectious Diseases
and Microbiology Translational Research, University of Minnesota,
Minneapolis, MN 55455, USA
PAULETTE M. ROBINSON Department of Molecular Genetics and Microbiology,
University of Florida, Gainesville, FL, USA
MENACHEM RUBINSTEIN Department of Molecular Genetics,
The Weizmann Institute of Science, Rehovot 76100, Israel
HUUB F.J. SAVELKOUL Cell Biology & Immunology group, Department of Animal
Sciences, Wageningen University,Wageningen PG 6709, Netherlands
GREGORY S. SCHULTZ Department of Obstetrics and Gynecology,
College of Medicine, Institute for Wound Research, University of Florida,
Gainesville, FL 32610-0294, USA
BRAD SHERMAN Laboratory of Immunopathogenesis and Bioinformatics,
MOTOMU SHIMAOKA Immune Disease Institute, Boston, MA, USA;
Program in Cellular and Molecular Medicine, Childrens Hospital Boston,
Boston, MA, USA; Department of Anesthesia, Harvard Medical School,Boston,
MA 02115, USA
CHARUDHARSHINI SRINIVASAN Immune Disease Institute, Boston, MA, USA;
Program in Cellular and Molecular Medicine, Childrens Hospital Boston,
Boston, MA, USA; Department of Anesthesia, Harvard Medical School,
Boston, MA 02115, USA
PETR SVOBODA Institute of Molecular Genetics, Academy of Sciences
JAN TAVERNIER Department of Medical Protein Research, Flanders Institute
for Biotechnology, Ghent University, Faculty of Medicine and Health Sciences,
Ghent BE-9000 , Belgium
Contributors
JOS VAN DER HEYDEN Department of Medical Protein Research, Flanders Institute
Ghent BE-9000, Belgium
GEERT F. WIEGERTJES Department of Animal Sciences, Cell Biology
& Immunology group, Wageningen University, Wageningen PG 6709,
Netherlands
SCOTT R. WITTING Department of Medical and Molecular Genetics,
and Center for Diabetes Research, Indiana University School of Medicine,
Indianapolis, IN 46202, USA
JUN YANG Laboratory of Immunopathogenesis and Bioinformatics, CSP,
ADD, SAIC-Frederick, Inc., NCI-Frederick, Frederick, MD 21702, USA
RONG YUAN Department of Obstetrics and Gynecology, College of Medicine,
Institute for Wound Research, University of Florida, Gainesville,
FL 32610-0294, USA
LENNART ZABEAU Department of Medical Protein Research, Flanders Institute
Ghent BE-9000, Belgium
JIRI ZAVADIL Department of Pathology, Center for Health Informatics
and Bioinformatics, New York University Langone Medical Center,
New York, NY 10016, USA
xi
Chapter 1
A Tale of Two Interferon Bioassays: How Frustration
with Discrepant Results from Slightly Dissimilar
Methods Can Engender Discovery
Alfons Billiau
Abstract
This introductory article describes an episode that took place in the mid-1980s when the first wave of
cytokine discoveries took place. During studies aimed at complete purification of human interferon- from
crude mitogen-stimulated lymphokine preparations, the use of two different antiviral bioassays for the
cytokine yielded disparate results. Analysis revealed the presence of a contaminant IFN-like cytokine that
was detectable with only one of the two assays. Superficially, the contaminant resembled IFN-. However,
further analysis showed that it was not an IFN at all but an IFN-inducing cytokine identifiable as
interleukin-1.
Key words: Bioassay, Interferon, Interleukin, Antiviral activity
1. Die Methode ist

Alles
This aphorism, attributed to Karl Friedrich Wilhelm Ludwig
(18161895), seems like an appropriate thought to reflect upon in
the introduction to this book. Witnessing, as we currently do, the
bewildering progress in the biological sciences, mainly enabled by
the booming of molecular gene technology, bioinformatics, and
imaging techniques, we hardly need the influential German physiologists statement to be reminded of the key role of methodology
in science in general. However, also in a more parochial context,
each one of us, at a certain point in his/her scientific career, experiences instances when a methodological detail, trivial as it may seem
at first, happens to provide the key to a scientific discovery.
Marc De Ley (ed.), Cytokine Protocols, Methods in Molecular Biology, vol. 820,
DOI 10.1007/978-1-61779-439-1_1, Springer Science+Business Media, LLC 2012
A. Billiau
This is exactly what happened to me and my coworkers (including

the Editor of this book, Marc De Ley) when, in the early 1980s,
we were working on the characterization of human interferons
(IFNs) in the Rega Institute of the University of Leuven, Belgium.
In a nutshell, what happened was that we became frustrated with a
discrepancy in the results from two only slightly different bioassays
for IFN-. Perseverance in trying to resolve these discrepancies led
us to the discovery of the now generally accepted cascade principle underlying the cytokine network, i.e., the fact that one
cytokine uses to induce production of several other ones. As it
happened, what we were able to show was that interleukin-1
induces production of IFN- and thereby can itself mimic IFN
(Fig. 1) (1). I already described this episode in great detail (2).
However, on request of the Editor, I here recall the main events.
IFNs became known mainly by their ability to reprogram cells
such that they mount a state of resistance to virus infection. Several
molecular types of IFN have been identified, some with subtypes.
The main types are IFN-, IFN-, and IFN-. IFN- and IFN-
share sequence homology and use the same cellular receptor; IFN-
is unrelated to all others and uses a different receptor. Besides
acting as inhibitors of viral replication in cells, IFNs have other
effects by which they contribute to inflammation and immune
Fig. 1. After stimulation with an appropriate mitogen, human peripheral blood mononuclear
cells produce various cytokines, including a 22-kDa protein possessing antiviral activity
that resembles HuIFN- in being neutralized by specific antiserum against this interferon.
In reality, 22 K is interleukin-1 endowed with the ability to induce IFN- production by
the diploid fibroblasts used in the interferon bioassay (reprinted with permission from
Immunol. Today, ref. 1).
A Tale of Two Interferon Bioassays
responses to infections and tumors. In the 1970s, our laboratory

had become strongly committed to the development of a system
for production of human IFN to be used for molecular characterization and clinical trials. As the cellular source of the IFN, we had
chosen to use serially subcultured human diploid skin fibroblasts
induced with synthetic, double-stranded RNA poly-rI:rC. This system
turned out to deliver the -type of human IFN, as opposed to the
-type which was produced by others who used human leukocytes
(collected from the buffy coats of blood donations) and stimulated
these by viral infection. In the process, we had acquired not only
the know-how for mass production of human IFN-, but we had
also developed various methodologies and reagents for bioassay,
purification, and characterization of the IFN. One reagent in particular, a potent antiserum against IFN- that did not cross-react
with other known IFNs, would become important in the events
that I wish to recall. Another important technicality was that, having
at our disposal lots of diploid human cell cultures, we used these
cells also for the bioassay of our IFN samples. Of note, today, most
laboratories working with IFNs use ELISAs to quantify their samples.
However, at the time, such assays were insufficiently developed. In
a classical IFN bioassay, cell cultures in microtiter plates are treated
with serial dilutions of the samples, and then infected with a challenge
virus. The highest dilution that still provides protection against the
virus gives a measure of the IFN content in a sample. Diploid cells
are reputed to be highly sensitive to the antiviral effect of all IFNs,
so they are a good choice in any case. Nevertheless, any other laboratory would tend to use a continuous cell line, such as Hep-2, as
these are much easier to keep in culture and to manipulate.
Around 1980, Marc De Ley (3) developed an interest in the
third major type of human IFN, IFN-, then still widely called
immune IFN, as it appeared to be produced only by lymphocytes and possessed strong immunoregulatory activities, e.g., macrophage activation. Today, IFN-, available as a laboratory reagent
from commercial suppliers, is usually a product of recombinant
DNA technology. At the time of my narrative, however, there was
no other choice than to produce IFN- by treating suspensions of
fresh leukocytes from blood donations with an appropriate mitogen,
mostly concanavalin A. De Leys primary objective was to obtain
sufficient IFN- in pure form to study its physical, chemical, and
biological properties. Two difficulties had to be overcome: low
yields in the production and a host of contaminants present in the
crude starting product. Batches of crude product were submitted
to a first rough concentration and purification step consisting of
adsorption to silicic acid and desorption by ethylene glycol. Pools
of this partially purified IFN were then further processed through
several additional steps, the first of which was gel filtration.
As soon as concentration and purification of the material began
to run more or less smoothly, I convinced another collaborator, Jo
Van Damme, who had acquired experience in mass production of
A. Billiau
human IFNs, to set up a routine of semi-mass production of IFN-

so as to provide De Ley with ample crude product for concentration and purification. Soon, a minor dispute arose as De Ley found
that samples seemed to contain less activity than expected from the
initial values communicated by Van Damme. It so happened, however, that my two collaborators were using different titration methods to determine the potency of the samples. Van Damme used
diploid fibroblasts as these were amply available in his laboratory;
De Ley used the easy-to-cultivate continuous human cell line
Hep-2 (alias HeLa or CCL-23). At one point, fractions of human
immune IFN separated by molecular mass were titrated on both
human cell types. It appeared that most of the antiviral activity
migrated in fractions of approximately 45 kDa (natural IFN-
proved later indeed to consist of a homodimer of 22 kDa glycoproteins). However, titration on diploid cells detected an additional
minor peak at 22 kDa, not revealed by titration on Hep-2 cells
(Fig. 2) (4). We became intrigued by this 22-K fraction and to
our surprise found that its antiviral activity was neutralized in the
Fig. 2. Elution profile on gel filtration of interferon produced by human peripheral blood mononuclear cells induced with
concanavalin A, prepurified by adsorption to silicic acid, and elution with ethylene glycol in 1.4 M NaCl. Solid circles: titration
by inhibition of viral infection using HEp-2 cells; empty circles: titration using human diploid cells (reprinted with permission
from Eur. J. Immunol., ref. 4).
A Tale of Two Interferon Bioassays
presence of antibodies against IFN-. We first speculated that 22 K

was an IFN- monomer that exposed internal epitopes shared with
IFN-. However, in experiments set up to purify 22 K by affinity
chromatography on the same anti-IFN- antibody, Van Damme
found that the activity percolated through the column without
binding. This led us to the hypothesis, ultimately found to be correct,
that 22 K was not an interferon but a different lymphokine-like
factor that exerted antiviral activity by inducing IFN- when
titrated on diploid fibroblasts but not on Hep-2 cells (5). Proving
this hypothesis turned out to be difficult as the amount of IFN
induced by 22 K was so small as to just suffice to act in a paracrine
fashion. It would take studies on mRNA expression using dot-blot
technology to obtain reasonably convincing evidence (6). Only as
late as 1989, with the advent of PCR technology, would induction
of IFN- (and other cytokines) by 22 K (meanwhile known to be
IL-1) became generally accepted (7).
However, independently from testing the IFN- induction
hypothesis, we wanted to know whether 22 K was a novel cytokine
or one that had already been described by others as being responsible for another biological activity. Solving this question involved
developing new bioassays and submitting our purified 22-K material to various laboratories with technological expertise in bioassays
for immunoregulatory or inflammatory factors, such as skinreactive factor (SRF), colony-stimulating factor (CSF), lymphotoxin,
lymphocyte-activating factor (LAF), monocytic cell factor (MCF),
and endogenous pyrogen (EP). These investigations revealed that
our highly pure 22-K factor had activities assigned to MCF (induction of collagenase and prostaglandin in synovial cell cultures) as
well as LAF activity, but not lymphotoxin activity. In addition, we
found that 22 K induced production of CSF in the same way as it
induced IFN- and thereby mimicked CSF in some assays (8).
These findings came at a time point when immunologists decided
to rename LAF and henceforth call it interleukin-1. When it became
evident that this cytokine was the active principle in various other
factors, including MCF, it dawned to us that 22 K was probably
identical to IL-1. We eventually succeeded in purifying the 22-K
factor to homogeneity and obtain sufficient material for determination of the amino-terminal amino acid sequence. It then appeared
that 22 K was nothing short of the mature form of interleukin-1,
as the sequence corresponded to part of an interleukin-1 cDNA
clone isolated at exactly the same time by molecular biologists in
the USA (9, 10).
The significance of these findings was twofold. They represented the first complete purification of natural interleukin-1 and,
by revealing the amino-terminal sequence, permitted to identify
the site of proteolytic cleavage of the precursor into the mature
interleukin-1 protein (11). Secondly, demonstration of IFN-
induction by an interleukin-1 represented one of the very first
A. Billiau
examples of a cytokine cascade. Induction of one cytokine by

another is now generally accepted as key to the regulatory function
of cytokine networks.
Today, thanks to gene and genome technology, most novel
bioactive proteins are first defined chemically before their biological activity becomes evident. However, at the time of my narrative,
the reverse pathway was the usual one: one would first detect a
peculiar activity in a biological fluid (serum, urine, cell culture
fluid, ) using a particular bioassay. Next, one would apply classical
protein-biochemical methods to concentrate, purify, and characterize the protein(s) responsible for the activity. Under these constraints on the process of discovery, the properties of the bioassay
were of key importance. What was true in Ludwigs time and in the
1980 when we were in the midst of the incipient wave of cytokine
discoveries remains true today. Today, as in the past, investigators
need to ascertain that their assays are highly sensitive and specific.
However, having a good understanding of the mechanistic chain of
events taking place during the assay is equally important.
One would hope that this book makes a substantial contribution to this goal.
References
1. Billiau, A., Opdenakker, G., Van Damme, J.,
De Ley, M., Volckaert, G., Van Beeumen, J.
(1985) Interleukin-1: amino acid sequencing
reveals microheterogeneity and relationship
with an interferon-inducing monokine.
Immunol. Today 6: 235236.
2. Billiau, A. (1987) The interferon-interleukin 1
connection. In Interferon, Academic Press
Inc., London, Vol. 9 pp.91111.
3. De Ley, M., Van Damme, J., Claeys, H.,
Weening, H., Heine, J.W., Billiau, A., Vermylen,
C., De Somer, P. (1980) Interferon induced in
human leukocytes by mitogens: production,
partial purification and characterization. Eur. J.
Immunol. 10: 877883.
4. Van Damme, J., De Ley, M., Claeys, H., Billiau,
A., Vermylen, C., De Somer, P. (1981)
Interferon induced in human leukocytes by
concanavalin A: isolation and characterization
of gamma and -type components. Eur. J.
Immunol. 11: 937942.
5. Van Damme, J., Billiau, A., De Ley, M., De
Somer, P. (1983) An interferon--like or interferon-inducing protein released by mitogenstimulated human leukocytes. J. Gen. Virol.
64: 18191822.
6. Van Damme, J., Opdenakker, G., Billiau, A.,
De Somer, P., De Wit, L., Poupart, P., Content, J.
(1985) Stimulation of fibroblast interferon
7.
8.
9.
10.
11.
production by a 22 K protein from human

leukocytes. J. Gen. Virol. 66: 693700.
Fujita, T., Reis, L.F., Watanabe, N., Kimura, Y.,
Taniguchi, T., Vilcek, J. (1989) Induction of
the transcription factor IRF-1 and interferonbeta mRNAs by cytokines and activators of
second-messenger pathways. Proc. Natl. Acad.
Sci. USA 86: 99369940.
Fibbe, W.E., Van Damme, J., Billiau, A., Voogt,
P.J., Duinkerken, N., Kluck, P.M.C., Falkenburg,
J.H.F. (1986) Interleukin-1 (22-K factor)
induces release of granulocyte-macrophage
colony-stimulating activity from human mononuclear phagocytes. Blood 68: 13161321.
Auron, P.E., Webb, A.C., Rosenwasser, L.J.,
Mucci, S.F., Rich, A., Wolff, S.M., Dinarello,
C.A. (1984) Nucleotide sequence of human
monocyte interleukin 1 precursor cDNA. Proc.
Natl. Acad. Sci. USA 81: 79077911.
Van Damme, J., De Ley, M., Opdenakker, G.,
Billiau, A., De Somer, P., Van Beeumen, J.
(1985) Homogeneous interferon-inducing
22 K factor is related to endogenous pyrogen
and interleukin-1. Nature 314: 266268.
Auron, P.E., Rosenwasser, L.J., Matsushima, K.,
Copeland, T., Dinarello, C.A., Oppenheim, J.J.,
Webb, A.C. (1985) Human and murine interleukin 1 possess sequence and structural similarities. J. Mol. Cell. Immunol. 2: 169177.
Chapter 2
The Use of Real-Time Quantitative PCR for the Analysis
of Cytokine mRNA Levels
Maria Forlenza, Thomas Kaiser, Huub F.J. Savelkoul,
and Geert F. Wiegertjes
Abstract
Over the last decade, real-time-quantitative PCR (RT-qPCR) analysis has become the method of choice
not only for quantitative and accurate measurement of mRNA expression levels, but also for sensitive
detection of rare or mutated DNA species in diagnostic research. RT-qPCR is based on the standard principles of PCR amplification in addition to the use of specific probes or intercalating fluorescence dyes. At
the end of every cycle, the intercalating dye binds to all double-stranded DNA. There is a quantitative
relationship between the amount of starting DNA and the amount of amplification product during the
exponential phase. However, to obtain meaningful RT-qPCR data, the quality of the starting material
(RNA, DNA) and the analysis method of choice are of crucial importance. In this chapter, we focus on the
details of RNA isolation and cDNA synthesis methods, on the application of RT-qPCR for measurements
of cytokine mRNA levels using Sybr-Green I as detection chemistry, and finally, we discuss the pros and
contras of the absolute quantification versus relative quantification analysis. RT-qPCR is a powerful tool,
but it should be handled with care.
Key words: Real-time-quantitative PCR, Absolute quantification, Relative quantification, Primer
efficiency, Housekeeping gene
1. Introduction
Over the last decade, real-time-quantitative PCR (RT-qPCR) analysis
has become the method of choice not only for quantitative and
accurate measurement of mRNA expression levels, but also for sensitive detection of rare or mutated DNA species in diagnostic
research (1, 2). RT-qPCR is based on the standard principles of
PCR amplification in addition to the use of specific probes or intercalating dyes. Various probe systems are available among which
M. Forlenza et al.
Fig. 1. A typical RT-qPCR profile can be divided in the initial, exponential, and plateau
phases.
TaqMan probes, Molecular Beacons, MGB probes, and others

increasing specificity and sensitivity of the real-time assays.
RT-qPCR, using intercalating dyes that become fluorescent upon
binding to double-stranded (ds) DNA, has the advantage of running melting curve analysis after each run in order to check specificity. In most cases, Sybr-Green I is used, but other dyes are
available, including Eva-Green, Syto9, etc.
Under optimal conditions, every PCR cycle should result in a
doubling of the amplification product. At the end of every cycle,
the intercalating dye binds to all double-stranded DNA. Ideally,
the increase in amount of template is directly proportional to the
increase in fluorescence. Fluorescence data are collected during
each cycle allowing for real-time monitoring of amplification.
A typical RT-qPCR profile is shown in Fig. 1: it can be divided in
the initial, exponential, and plateau phases. The exponential phase
of the amplification provides the most useful and reproducible
data. There is a quantitative relationship between the amount of
starting DNA and the amount of amplification product during the
exponential phase.
The number of cycles required for a sample to rise above the
background fluorescence and reach the threshold level is called
Ct-value (threshold cycle). The threshold is set at a level, where the
rate of amplification is greatest during the exponential phase, allowing for the most accurate and reproducible results. An advantage of
RT-qPCR over conventional PCR is the possibility to assess the
amplification efficiency (E). Particularly, when the expression profile of more genes needs to be compared, it is important to take the
efficiency into account and adjust for differences between different
genes to be compared. In addition, at the end of every run, a melting curve analysis can be performed to assess amplification specificity. Taken together, this leads to increased sensitivity, specificity,
and efficiency of the PCR analysis. To obtain meaningful RT-qPCR
data, the quality of the starting material (RNA, DNA) and the
analysis method of choice are of crucial importance. In this chapter,
The Use of Real-Time Quantitative PCR for the Analysis of Cytokine mRNA Levels
we focus on the details of RNA isolation and cDNA synthesis methods,

the application of RT-qPCR for measurements of cytokine mRNA
levels using Sybr-Green I as detection chemistry, and finally, we discuss the pro and contras of the absolute quantification versus relative quantification analysis.
1.1. Absolute
Quantification
Absolute quantification analysis ideally determines the absolute

copy number of a gene of interest (GOI) in an unknown sample.
The unknown sample is compared to a standard curve with known
concentrations of template. In most cases, recombinant plasmid
DNA (recDNA), cDNA, recRNA, pooled samples, or PCR products are used for this purpose. Therefore, the accuracy of the absolute quantification assay entirely depends on the accuracy of the
standard (3). No matter how accurate the concentration of the
standard material is the final result is always expressed relatively to
a defined unit of interest: e.g. copies per ng of total RNA, copies
per cell, copies per gram of tissue, copies per mL blood. When
absolute changes in copy numbers are important, the denominator has to be shown to be absolutely stable across the comparison.
Although the word absolute suggests an exact measurement,
one has to be aware that absolute quantification is relative to the
standards used.
1.2. Relative
Quantification
Relative quantification analysis determines the levels of expression

of a GOI and expresses it relative to the levels of an internal control
or reference gene (RG). Results are given as ratio of GOI versus
one or more RGs (4). In this type of analysis, the function of the
RG is to normalize the data for differences in RNA (DNA) quantification and template input. Therefore, expression of the RG has
to be analyzed in the same sample as the GOI and can be coamplified in the same tube as a multiplex assay (probes) or the
same sample should be used in separate tubes as a simplex assay
(Sybr-Green I).
Reference genes are genes that are not affected by the treatment in any way and are constant under the tested conditions.
Hence, the reliability of the relative quantification analysis is strongly
dependent on the stability of the RG. Several tools are available for
the determination of the best RG: TATA Biocenter AB: http://
www.tataa.com/Products/Human-Endogenous-Control-Panel.
html; geNorm (5): http://medgen.ugent.be/~jvdesomp/
genorm/; BestKeeper (6): http://www.gene-quantification.info.
We have extensive experience with the BestKeeper software.
1.3. Real-Time PCR

Cyclers
Most RT-qPCR cyclers make use of a solid block (96 or 384 wells)
for thermal cycling while others use hot and cooled air. Most of the
solid block-based real-time instruments are affected by thermal
variation across the block and by differences in illumination and
10
M. Forlenza et al.
optical signals detected from each sample. Both aspects greatly

contribute to well-to-well variability. Two air-based cyclers employ
a rotary design using capillaries or plastic tubes and one of them
uses a centrifuge, which guarantees optimal thermal and optical
uniformity. Samples are continuously rotating in the thermal chamber, guaranteeing minimal temperature variation between tubes in
contrast to positional effects, such as the recognized edge effect
observed in block-based designs. In addition, every tube moves
past the identical excitation light source and detection pathway,
which guarantees optical uniformity. In our laboratory, we have
extensive experience with the Rotor-Gene 6000.
2. Materials
2.1. RNA Isolation
and cDNA Synthesis
1. RNA isolation, including on-column DNase treatment: RNeasy

Mini Kit and RNase-free DNase set (QIAgen).
2. cDNA synthesis, including DNase treatment: DNase I,
Amplification Grade; Superscript III First Strand Synthesis
Systems for RT-PCR Systems (Invitrogen).
3. Nuclease-free water (Promega).
4. RT-qPCR Master mix: ABsolute QPCR SYBR Green Mix
(ABgene).
5. NanoDrop spectrophotometer (Thermo Scientific).
6. Thermal cycler: Rotor-Gene 6000 (Corbett Research).
More detailed information to any RT-qPCR topic can be found
on the following Web site: http://www.gene-quantification.info
2.2. Plasmid
Construction
and Isolation
1. Luria Bertani (LB) medium (1 L).

2. LB plates.
3. E. coli JM109 High Efficiency Competent Cells (Promega).
4. pGEM-T easy Ligation Kit (Promega).
5. QIA prep Spin Miniprep kit (QIAgen).
6. Gel Extraction Kit (QIAgen).
7. Ampicillin.
8. X-gal.
9. IPTG.
10. SOC medium.
11. 100% glycerol.
12. Agarose.
11
3. Methods
3.1. RNA Isolation
and Quantification
Isolation and quantification of good-quality RNA (see Note 1) are

of extreme importance to obtain meaningful gene expression data
by RT-qPCR. Several commercial kits are available; for RNA isolation,
from small (30 mg), fresh-frozen or RNA-later stored tissue samples and from primary cells or cell lines (107 cells), we obtained
high-quality results with the RNeasy mini kit from Qiagen.
1. Isolate RNA according to the manufacturers instructions.
Work fast, clean, wear gloves, and use RNase-free tubes and
tips. To reduce genomic DNA (gDNA) contaminations,
include an on-column DNase digestion step. Elute RNA in
3050 L RNase-free water.
2. Use 12 L of the eluted sample to determine RNA concentration (OD measurement at 260 nm) and RNA quality (OD
260/280 ratio) with the NanoDrop spectrophotometer. An
OD 260/280 ratio greater than 1.8 is usually considered an
acceptable indicator of good RNA quality. The presence of
gDNA in the sample leads to an overestimation of the RNA
concentration.
3. RNA integrity and the absence of gDNA can be assessed by
loading 12 L of RNA sample on a 1% agarose gel. Two
major bands corresponding to the 28S and 18S rRNA should
be clearly visible. In case of gDNA contaminations, an additional band of higher molecular weight than the two rRNA
bands can be observed.
3.2. cDNA Synthesis
Several kits are available for cDNA synthesis. We routinely use the
SuperScript III First strand cDNA synthesis kit with random
primers from Invitrogen.
1. Prior to cDNA synthesis from 1 g of total RNA (see Note 2),
perform a second DNase digestion step using the DNase I
Amplification Grade Kit (Invitrogen).
2. Proceed with the cDNA synthesis protocol according to the
manufacturers instructions. For each sample, always include a
control for gDNA contaminations: in this sample, the same
amount of RNA is used but no reverse transcriptase is added to
the mix (RT control).
3. After cDNA synthesis, the final volume for each sample is
20 L. We routinely bring the volume up to 100 L and consider this our stock sample solution. Depending on the organ
or cell type, we further dilute the stock five to ten times. This
allows performing up to 200 reactions for each sample when
using 5 L of template in each PCR reaction.
12
M. Forlenza et al.
3.3. Construction of
Recombinant Plasmid
DNA
The calibration curves used in absolute quantification can be based

on known concentrations of DNA standard molecules, e.g.
recDNA, gDNA, RT-PCR product, commercially synthesized, big
oligonucleotide (see Note 3). In this section, we describe how to
construct a recombinant plasmid DNA containing the sequence of
any GOI.
1. Design primers to amplify a large (5001,000 bp) fragment of
the gene. The region should of course contain the sequence to
which the primers designed for RT-qPCR anneal. Amplify the
large product by conventional PCR or reverse transcriptasePCR.
2. Gel purify the product using the QIAgen Gel Extraction Kit
and elute in 30 L of water.
3. Ligate the product into the vector by combining 3.5 L of the
gel-purified product to 5 L of 2 ligation buffer, 0.5 L
(25 ng) of pGEM-T easy, and 1 L (3 UI) of T4 DNA ligase
(all reagents in the easy ligation kit). Mix by pipetting, and
incubate for 1 h at room temperature or overnight at 4C for
the maximum number of transformant.
3.4. Amplification
and Quantification
of recDNA
1. Prepare LB agar plates containing ampicillin, X-Gal, and

IPTG.
2. Centrifuge the ligation reactions briefly. Add 25 L of each
ligation reaction to a sterile 10-mL tube on ice.
3. Thaw one vial (200 L) of JM109 High Efficiency Competent
Cells on ice. When just thawed, mix the cells by gently flicking the
tube. Carefully transfer 50 L of cells to the ligation tube from
step 2. Gently flick the tube and incubate on ice for 20 min.
4. Heat shock the cells for 4550 s in water bath at exactly 42C.
DO NOT SHAKE. Immediately return the tube to ice for
2 min.
5. Add 950 L room temperature SOC medium to each reaction
tube. Incubate for 1.5 h at 37C with shaking (~150 rpm).
6. Transfer the total volume of the transformation reaction to an
Eppendorf tube, and centrifuge for 10 min at 350 g. Remove
900 L of medium and resuspend the bacterial pellet in the
remaining 100 L. Spread 90 and 10 L of cell suspension
onto two LB agar plates containing ampicillin, X-Gal, and
IPTG and incubate overnight at 37C.
7. With a sterile pipette tip, tick pick 58 white colonies and
transfer each of them in 4 mL LB medium containing ampicillin (50 g/mL). Grow overnight with shaking at 300 rpm.
8. Isolate plasmid from 3 mL of the overnight culture using the
QIAgen QIA prep Spin Miniprep kit. Elute plasmid in 50 L
of water.
13
9. Make glycerol stocks by combining the remaining 1-mL overnight

culture to 200 L 100% glycerol.
10. Load 12 L of isolated plasmid on a 1% agarose gel. Three
bands of high molecular weight corresponding to the linear, circular, and supercoiled forms of the plasmid should be visible.
11. Linearize the plasmid by combining 30 L of purified plasmid
to 3 L of restriction enzyme of choice, 5 L of the appropriate
10 reaction buffer, and water up to a final volume of 50 L.
12. Gel purify the linearized plasmid using the QIAgen Gel
Extraction Kit and elute in 30 L of water.
13. Determine plasmid concentration using the NanoDrop. Take
an average out of at least five measurements (better ten) and
perform the measurement at multiple template dilutions. The
concentration of the plasmid has to be calculated very accurately
because this measurement determines the outcome of the absolute
quantification analysis. For use in RT-qPCR, prepare the plasmid as described below.
3.5. Calculation
of Plasmid Copy
Number
and Preparation
of the Standard Curve
Once the size of the plasmid containing the GOI is known, it is

possible to calculate the number of grams/molecule, also known
as copy number, as in the following example:
Weight in Daltons (g/mol) = (bp size of plasmid + insert)(330
Da 2 nucleotides/bp).
Ex. g/mol = (5,950 bp)(330 Da 2 nucleotides/bp) = 3,927,000
g/mol.
Hence, (g/mol)/Avogadros
g/molecule = copy number.
number
6.02214199 1023 =
Ex. 3,927,000 (g/mol)/6.02214199 1023 = 6.52 1018 g/molecule.

The precise number of molecules can be determined as follows:
Concentration of plasmid (g/L)/copy number.
Ex.
(3 107
molecules/L.
g/L)/(6.52 1018g/molecule) = 4.6 1010
Once the number of molecules in 1 L of linearized plasmid

solution is calculated, prepare standard dilutions to obtain an X
plasmid copy number in 5 L of water. Accurate pipetting is essential because the standards must be diluted over several orders of
magnitude. It is recommended to divide standards into small aliquots, store at 80C, and thaw only once before use.
3.6. Primer Design
The design of specific primers that work at a good efficiency is of

crucial importance in RT-qPCR. Use the Primer3 program
(http://frodo.wi.mit.edu/cgi-bin/primer3/primer3_www.cgi) to
design primers of a length of 1822 bp, with an annealing
temperature of 60C and a minimum self and 3 complementarily.
To accomplish rapid quantification, short PCR cycling (4575 s),
14
M. Forlenza et al.
Fig. 2. Melting curve profile of PCR products amplified with three different primer sets.
Light grey: Four PCR products each showing the same specific melting peak. Dark grey :
Four PCR products of which three showing a specific melting peak and a fourth one being
a non-specific amplification product with a different melting temperature. In black:
Amplification with the third set of primers resulted only in primer dimer formation.
and efficient PCR conditions, the optimal length of the PCR product
is 100200 bp. Use the OligoAnalyzer program (http://eu.idtdna.
com/analyzer/Applications/OligoAnalyzer/Default.aspx?c=EU)
to verify that the primers have low self- and hetero-complementarity. To increase the annealing temperature of primers, to improve
the specificity of allele-specific primers, or for single-nucleotide
polymorphism (SNP) analysis, the incorporation of locked nucleic
acid (LNA) modifications can be of great advantage (7, 8). Software
program to estimate melt behaviours of a template is POLAND
MELTSIM (http://www.bioinformatics.org/meltsim/wiki/).
3.7. PCR Profile
and Melting Curve
Analysis
A typical PCR profile includes an initial denaturation step of

1015 min at 95C, depending on the Taq-Polymerase (see Note 4),
followed by 3540 cycles, including 95C for 515 s (denaturation), 60C for 1530 s (annealing), and 72C for 1530 s (elongation). This profile is a general suggestion and the annealing
temperature has to be verified. At the end of the run, a melting
step needs to be performed to assess amplification specificity (see
Note 5, Fig. 2). Each PCR product has a specific melting temperature, resulting in a single melting peak with no additional peaks at
lower melting temperatures. Additional peaks can be primer dimers or unspecific products due to excessive amount of primers in
the reaction, low annealing temperature, too high MgCl2 concentration, or too long hold times. Primer dimers formation can be
reduced or eliminated by accurate design of the primers and optimization of primer concentration. When using a primer set for the
first time, despite the presence of only one amplification peak, it is
advised to sequence at least once the amplification product to confirm sequence specificity.
3.8. Optimization
of Primer
Concentration
Select a cDNA template or recDNA containing the sequence of the

GOI. Prepare a master mix containing 7 L of 2 Sybr-Green I
Mix and 5 L of DNA. Aliquot 12 L of the master mix into reaction
15
tubes and add 1 L of each primer to give final concentrations as

outlined in the table below. The final reaction volume is 14 L (see
Note 6). The primer stock concentrations are 1.4, 4.2, and 7 M
and give final concentrations of 100, 300, and 500 nM,
respectively.
Forward primer (nM)
Reverse primer (nM)
100
300
500
100
100/100
300/100
500/100
300
100/300
300/300
500/300
500
100/500
300/500
500/500
We usually find 300 nM the optimal concentration for both

forward and reverse primers. As a general guideline, choose the
primer combination which gives the lowest Ct value for the same
amount of template and does not lead to primer dimer formation.
In every run, always include a non-template control (NTC), where
the template is replaced by the same amount of water, in order to
test for primer specificity and contaminations.
3.9. Determination
of Primer
Amplification
Efficiency
Depending on the subsequent method of analysis, there are several

ways to determine primer amplification efficiencies. The most commonly used is the standard curve method: a dilution series of a
reference template or pooled samples of unknown concentration is
generated. The reference sample can be cDNA or recDNA (of
unknown concentration) that contains the target gene. The units
used to describe the concentration of the dilution series are relative, as long as they reflect the dilution factor of the standard curve
(Fig. 3).
Set the threshold just above the take-off point of the reactions
(if the result for more genes over different experiments need to be
compared, set the threshold at the same level for all genes, for example 0.1). Record the Ct values and plot them against the log template concentration. Use the slope of the regression line to calculate
the amplification efficiency for each primer according to the following formula: E = 10 (1/slope). The optimal amplification efficiency of a
reaction is 2, but we consider E values between 1.7 and 2 as acceptable, as long as the reproducibility over several runs as well as the
replicates is good. Usually, all RT-qPCR software provide this type
of calculations (see example in Table 1 and Fig. 4). In general, it is
important that the amplification efficiency of the reference template
reflects the amplification efficiency of the unknown sample.
3.10. Relative
Quantification
Analysis
Relative quantification is the method of choice for RT-qPCR analysis when investigating physiological changes in gene expression
levels. It does not require standard curves with known concentration
of templates and results are given as the ratio (R) of GOI versus
16
M. Forlenza et al.
Fig. 3. Standard curve of a tenfold dilution series of a reference cDNA sample used to
calculate the amplification efficiency of the primer sets for the RG and GOI.
Table 1
Results obtained from the RG standard curve described in Fig. 3. By plotting the
averaged Ct values from duplicate samples against the log of the given concentration,
the corresponding standard curve will be obtained as shown in this table.
17
Fig. 4. Possible RT-qPCR set up for relative quantification of IL-10 mRNA expression levels.
A first run, where a standard curve for each of the analyzed genes is amplified has to be
performed when analyzing the data using DDCt and Pfaffl method (see text). Particularly,
for the validation experiment required for the DDCt method, it is important the template to
be the same for each gene which needs to be compared. In a second run (the experimental run), amplify the RG (40S) and the GOI (IL-10) in each of the samples under investigation. Always include a non-template control (NTC), where water substitutes the template,
and a control for genomic contamination (RT). When a standard curve should be imported
from a previous run, include a triplicate sample of one dilution point of the same standard
curve in the current run (see Note 7).
one or more RGs. To date, several mathematical models have been

developed and can be generally divided into two major categories:
without and with primer efficiency correction. In this section, we
provide examples on how to analyze an experiment applying both
of those methods. For example, we want to determine the fold
change in interleukin-10 (IL-10) mRNA expression at various time
points after treatment. The 40S ribosomal protein S11 is used as
RG. Set up the first run of the day by amplifying a standard curve
for each target gene using a template cDNA or recDNA of unknown
concentration (Fig. 4; it might not be necessary to run a standard
curve every time depending on the chosen method of analysis). Set
up a second run, where in separate tubes the 40S and IL-10 genes
are amplified for each of the samples under investigation. Include a
triplicate sample of one dilution point of the same standard from
the first run. Analyze the results according to one of the methods
outlined below.
3.10.1. Relative
Quantification Without
Efficiency Correction: DDCt
Method
The Ct method (9) is based on the assumption that the primers

of the GOI will have the same amplification efficiency as the primers for the RG. This assumption needs to be validated at least once
before proceeding with the analysis of the experiment. See Note 8
for instructions on the validation experiment. In case of positive
results from the validation experiment, proceed as follows.
Set the threshold to 0.1 for all genes.
Export the Ct values to Excel.
18
M. Forlenza et al.
Select the sample at time point 0 (zero) as the calibrator (the

calibrator is usually an untreated, unhandled sample).
Apply the following formula:

DDCt = (Ct IL -10 - Ct 40S )(sample) - (Ct IL -10 - Ct 40S )(calibrator) ,
R(IL - 10) = 2- DDCt.
This method has the advantage that standard curves are

required only once for the validation experiment and allows for
normalization relative to an internal reference gene. However, the
assumption that different primer sets will perform with the same
amplification efficiency over different runs and over different templates might not always be valid. Therefore, the efficiency of all RG
and GOI should be checked regularly, as changes in reagents, concentrations, calibrator, etc. could influence the efficiency of one or
various genes differently.
3.10.2. Relative
Quantification with
Efficiency Correction:
The Pfaffl Method
This method does not require the amplification efficiency of different

primer sets to be similar; it rather takes into account the possibility
that the efficiencies can be different and offers a way to correct for
such differences (4, 10). Optimally, a standard curve for each of
the target genes is amplified in the same run together with the
unknown samples. However, when a large number of samples and
numerous genes need to be analyzed, standard curves for several
genes can be amplified in the first run of the day or even on a different day (Fig. 4).
Set the same threshold for all genes to be analyzed (i.e. 0.1)
and record the amplification efficiency for each primer set as
described in the previous paragraph.
In the experimental run, it is possible to either import the standard curve from the previous run and ask the software to adjust
it to the standard in the current run (Run 2 in Fig. 4) or the
threshold can be directly set manually to 0.1.
Export the Ct values to Excel.
Select the sample at time point 0 (zero) as calibrator and apply

the following formula:
(Ct
RIL -10 =
-10(calibrator)
E (IL -IL10)
(Ct
40 S(calibrator)
E E (40S)
- Ct IL -10(sample1-7) )
- Ct 40 S(sample1-7) )
The Pfaffl method is a modification of the Ct method with

the obvious advantage that it does take into account differences in
amplification efficiencies between primer sets.
In order to obtain direct and valuable statistical information, it
is possible to import the above-mentioned data in the gene
19
quantification software called Relative Expression Software Tool

(REST, freely available at http://rest.gene-quantification.info).
This software uses the Pfaffl formula and generates statistical data,
including the standard error and the confidence interval, by using
randomisation tests via hypothesis testing P(H1) = difference
between sample and control is due only to chance.
3.10.3. Relative
Quantification with
Efficiency Correction:
Sigmoidal or Logistic
Curve Fitting Models
To date, several methods have been developed to calculate the

amplification efficiency of each primer set in each single sample
(1113). The great advantage of all these methods is that they do
not require the preparation of standard curves or validation experiments and no assumption has to be made regarding the amplification efficiency of each primer set over different runs, templates, or
master mixes. The method developed by Corbett Research has
been incorporated in the Rotor-Gene 6000 software under the
Comparative Quantitation analysis option and we routinely use
it for our relative quantification of gene expression. We directly
apply the set up of Run 2 in Fig. 4 (without the need of the standard samples).
The Ct values and the amplification efficiency for each sample

are directly obtained from the software and exported to
Excel.
The average amplification efficiency (EA) for each primer in

each run is calculated and the relative fold change for each
GOI is calculated according to the Pfaffl formula as above.
It often happens that the analysis of one large experiment cannot be completed within one run. In that case, we calculate the
EA of each primer set over the whole experiment (two, three,
or more runs). To reduce variation between runs, we usually
prepare one master mix for each primer set which is enough for
all runs of the day and not one master mix for each run. By
doing so, we observe only a 0.02 variation in EA for each
primer set between two, three, or more runs on a single day.
Before using a new primer set for the first time, we perform a
dilution series of a cDNA sample containing the target gene. This
provides us with an estimation of the amplification efficiency and
the melting curve analysis provides us the specificity of the assay.
3.11. Absolute
Quantification
Analysis: External
Standard Curve Model
Absolute quantification refers to an analysis, where unknown samples

are compared to a standard curve of cDNA, recDNA, or recRNA,
where the absolute concentration is known. Especially for absolute
quantification analysis, the standard curve for the target gene
should be amplified in the same run together with the unknown
samples. However, when a large number of samples and numerous
genes need to be analyzed, it is possible to import a standard curve
from a previous run.
20
M. Forlenza et al.
Standard curves for several GOI can be amplified in the first

run of the day and in every subsequent run, together with the
unknown samples. A triplicate of one dilution point of the
standard curve should be included.
At the end of the run, ask the software to import the standard
curve for the GOI from a previous run and adjust it to the
standard in the current run (see Note 7). Read the absolute
copy number given by the software.
Alternatively, it is possible to export data to excel and perform

the quantification analysis by plotting the Ct values of the
unknown sample against the standard line obtained by plotting
the Ct values and the log concentration of the recDNA as
described before.
Express data as GOI (copy number)/x ng total RNA.
To normalize data and correct for variations in template input,

a normalizer (RG) is used. In this way, the absolute copy number of an RG and GOI in an unknown sample is determined
from the standard curve. The absolute value obtained for the
GOI is divided by the absolute value obtained for the RG in
the same sample. Obtained are the normalized data of the GOI
in the unknown sample. The quality of the gene quantification
data cannot be better than the quality of the denominator. Any
variation in the denominator obscures real changes, produces
artificial changes, and wrongs quantification results.
When optimized, standard curves are highly reproducible and

allow the generation of highly specific, sensitive, and reproducible
data. However, the external standard curve model has to be thoroughly validated as the accuracy of absolute quantification in realtime reverse transcriptase-PCR depends entirely on the accuracy of
the standards. Standard design, production, determination of the
exact standard concentration, and stability over long storage time
are not straightforward and can be problematic.
3.12. Technical or
Biological Replicates?
Depending on the applications, the use of technical and biological

replicates or both has to be considered. A technical replicate refers
to a sample, for example a piece of tissue, from which the RNA
isolation and cDNA synthesis has been performed more than one
time under the same identical conditions. This type of replicate
tells us something about the variation in the chemistry we are
using. Often, the same cDNA sample is analyzed in triplicate in
one RT-qPCR run. This type of technical replicate only tells something about the pipetting skills of the operator and the accuracy of
the PCR instrument (see also Subheading 1.3), but should absolutely NOT be considered for statistical analysis. Biological replicates refer to the application of the same treatment to two or more
samples. From each of the samples, the RNA isolation and cDNA
21
synthesis are performed independently but under identical conditions.

Each of the obtained cDNA samples can be analyzed once by
RT-qPCR.
Both types of replicates (technical or biological) provide
information about the experimental variation and allow statistics
to be applied to identify differences in expression levels between
samples. Being a beginner, it is a good practice to include technical
replicates to test for pipetting skills. When testing the amplification
efficiency of a new primer set, it is advisable to include at least a
triplicate of each dilution point. When investigating the effects of a
treatment, the use of biological replicates we think is of greater
value (14).
For example, in an in vitro experiment, cells are incubated in
the presence or absence of a stimulus. The treatment is repeated in
at least three replicate wells. Each of the three wells is a biological
replicate; however, the cells are derived from a single individual.
More relevant would be to repeat the same in vitro experiment on
cells isolated from three different individuals, each of them being a
biological replicate.
4. Notes
1. The extraction and purification procedure of total RNA must
fulfill the following criteria: free of protein (absorbance
260/280 nm); free of genomic DNA; should be non-degraded
(28S:18S ratio should be roughly between 1.8 and 2.0, with
low amount of short fragments); free of enzymatic inhibitors
for RT and PCR reaction, which is strongly dependent on the
purification and clean-up methods; free of any substances
which complex essential reaction cofactors, like Mg2+ or Mn2+;
free of nucleases for extended storage (15).
2. From 0.1 ng up to 5 g, total RNA can be transcribed into
cDNA using this kit. Optimally, 1 g of total RNA is used. In
general, it is important to use the same amount of starting
RNA material for each sample within the same experiment.
This greatly reduces the sample-to-sample variation due to differences in cDNA synthesis efficiency and simplifies the subsequent analysis, particularly when absolute quantification is
used. In some cases, not all samples (within the same experiment) would yield RNA amounts sufficient to use 1 g of
RNA/sample; it is possible to lower the amounts down to
0.1 g, but again this amount should be used for all samples
within the same experiment.
3. Cloned recDNA and gDNA are very stable and generate highly
reproducible standard curves even after a long storage time.
22
M. Forlenza et al.
Furthermore, the longer templates derived from recDNA and

gDNA mimic the average native mRNA length of about 2 kb
better than shorter templates derived from RT-PCR product
or oligonucleotides. A problem with DNA-based calibration
curves is that they are subject to the PCR step only, unlike the
unknown mRNA samples that must first be reverse transcribed.
This increases the potential for variability of the RT-PCR
results and the amplification results may not be strictly comparable with the results from the unknown samples (3).
4. The initial denaturation time depends on the type of TaqPolymerase present in the master mix. We strongly advise HotStart Taq-Polymerases that require 2 to 15 min at 95C,
depending on the Taq-Polymerase. This allows performing the
preparation and aliquoting of the master mix on the bench at
room temperature.
5. At the end of a run, after the last annealing step, all amplification products are present as double-stranded DNA and Sybr
Green I is bound to it. During the melting step, the decrease
in fluorescence is measured due to melting of dsDNA products
and consequent release of the fluorescent dye. Each product
melts at a specific temperature. Primer dimers usually have a
lower melting temperature than PCR products ranging
between 80 and 200 bp.
6. Usually, companies advise a final volume of 50 L, but the
reaction can easily be scaled down to save costs. We always try
to add at least 5 L of template. Lower volumes might increase
the chance of pipetting errors.
7. The slope of the calibration curve is more reproducible than
the intercept, and the slope directly correlates with PCR efficiency. Hence, only a single standard point is required to reregister a previously performed calibration curve level for the
new unknown samples. However, this assumes that the efficiency in a given run is the same as in a previous run.
8. Amplify a standard curve as described in the Subheading 3.9.
In this case, the reference template has to be the same for both
primer sets, and preferably one of the cDNA samples which is
going to be used for the subsequent experiment.
After having set the threshold (0.1), export the Ct values

to EXCEL and average the Ct of replicate samples.
Calculate the LOG10 of the given arbitrary concentration

(LOGconc).
Obtain the DCt: for each dilution point, calculate the difference between the Ct(RG) and Ct(GOI). Plot the LOGconc
vs. Ct and obtain the equation of the curve.
23
If the efficiencies of the two primer sets are approximately

equal, the obtained curve should be a nearly horizontal line with a
slope <0.1. If this is the case, the experiment can be analyzed with
the DDCt method.
References
1. Nolan, T., Hands, R.E., Bustin, S.A. (2006)
Quantification of mRNA using real-time
RT-PCR. Nat. Protoc. 1: 15591582.
2. Bustin, S.A., Benes, V., Nolan, T., Pfaffl, M.W.
(2005) Quantitative real-time RT-PCR - a perspective. J. Mol. Endocrinol. 34: 597601.
3. Pfaffl, M.W., Hageleit, M. (2001) Validities of
mRNA quantification using recombinant RNA
and recombinant DNA external calibration
curves in real-time RT-PCR. Biotechnol. Lett.
23: 275282.
4. Pfaffl, M.W. (2001) A new mathematical model
for relative quantification in real-time RT-PCR.
Nucleic Acids Res. 29: e45.
5. Vandesompele, J., De Preter, K., Pattyn, F.,
Poppe, B., Van Roy, N., De Paepe, A.,
Speleman, F. (2002) Accurate normalization of
real-time quantitative RT-PCR data by geometric averaging of multiple internal control
genes. Genome Biol. 3: RESEARCH0034.
6. Pfaffl, M.W., Tichopad, A., Prgomet, C.,
Neuvians, T.P. (2004) Determination of stable
housekeeping genes, differentially regulated
target genes and sample integrity: BestKeeper-Excel-based tool using pair-wise correlations.
Biotechnol. Lett. 26: 509515.
7. Latorra, D., Arar, K., Hurley, J.M. (2003)
Design considerations and effects of LNA in
PCR primers. Mol. Cell. Probes 17: 253259.
8. Latorra, D., Campbell, K., Wolter, A., Hurley,
J.M. (2003) Enhanced allele-specific PCR discrimination in SNP genotyping using 3 locked
nucleic acid (LNA) primers. Hum. Mutat. 22:
7985.
9. Livak, K.J., Schmittgen, T.D. (2001) Analysis

of Relative Gene Expression Data Using RealTime Quantitative PCR and the 2-[Delta]
[Delta]CT Method. Methods 25: 402408.
10. Pfaffl, M.W., Horgan, G.W., Dempfle, L.
(2002) Relative expression software tool
(REST) for group-wise comparison and statistical analysis of relative expression results in
real-time PCR. Nucleic Acids Res. 30: e36.
11. Liu, W., Saint, D.A. (2002) A new quantitative
method of real time reverse transcription polymerase chain reaction assay based on simulation of polymerase chain reaction kinetics.
Anal. Biochem. 302: 5259.
12. Ramakers, C., Ruijter, J.M., Deprez, R.H.,
Moorman, A.F. (2003) Assumption-free analysis of quantitative real-time polymerase chain
reaction (PCR) data. Neurosci. Lett. 339:
6266.
13. Tichopad, A., Dilger, M., Schwarz, G., Pfaffl,
M.W. (2003) Standardized determination of
real-time PCR efficiency from a single reaction
set-up. Nucleic Acids Res. 31: e122.
14. Forlenza, M., de Carvalho Dias, J.D., Vesely,
T., Pokorova, D., Savelkoul, H.F., Wiegertjes,
G.F. (2008) Transcription of signal-3 cytokines, IL-12 and IFN alpha beta, coincides with
the timing of CD8 alpha beta up-regulation
during viral infection of common carp
(Cyprinus carpio L). Mol. Immunol. 45:
15311547.
15. Fleige, S., Pfaffl, M.W. (2006) RNA integrity
and the effect on the real-time qRT-PCR performance. Mol. Asp. Med. 27: 126139.
Chapter 3
Interleukin-27 Induces Interferon-Inducible Genes:
Analysis of Gene Expression Profiles Using Affymetrix
Microarray and DAVID
Tomozumi Imamichi, Jun Yang, Da Wei Huang, Brad Sherman,
and Richard A. Lempicki
Abstract
We have previously demonstrated that IL-27 is a novel anti-HIV cytokine, which inhibits HIV replication
in CD4 T cells and macrophages as interferon (IFN)- does. To further understand the mechanism of the
antiviral effect, we performed Affymetrix DNA microarray and gene functional annotation analysis using
DAVID (the Database for Annotation, Visualization, and Integrated Discovery). DAVID is a web-based
bioinformatics application that systematically identifies enriched biology associated with large gene list(s)
derived from high-throughput genomic experiments, such as microarray. The enriched annotation terms
identified by DAVID will give important insights into understanding the biological themes under study.
Having used the DAVID bioinformatics tools, we have shown that IL-27 differentially regulates the gene
expression between T cells and macrophages. IL-27 significantly induces IFN-inducible genes including
antiviral genes in macrophages as does IFN-, suggesting that IL-27 inhibits HIV replication in macrophages
via a mechanism similar to that of IFN-.
Key words: IL-27, IFN-inducible genes, Microarray, DAVID, T cells, Macrophages
1. Introduction
IL-27 is a member of the IL-12 family of cytokines that consists
of IL-27p28 (also known as IL-30) and EpsteinBarr virus
induced gene 3(EBI3) (1). The p28 chain is related to IL-12p35
and has a classical cytokine structure, while EBI-3 is related with
IL-12p40 and structurally resembles the soluble IL-6 receptor
alpha chain. IL-27 binds to its receptor, IL-27R which is composed of a ligand-specific chain. IL-27R [Wsx1, T cell cytokine
25
26
T. Imamichi et al.
receptor (TCCR)] belonged to Type I Cytokine receptor family,

and gp130, a signal-transducing molecule shared with other
cytokines, IL-6, IL-11, and LIF. IL-27 is capable of binding to
IL-27R in the absence of gp130. However, the co-expression of both
receptor subunits is required to induce signal. Upon ligand binding, phosphorylation of STAT-1, -2, -3, -4, -5, or -6 occurs (2).
Despite the fact that IL-12 activates HIV-1 replication (3), we
have demonstrated that IL-27 is a potent anti-HIV cytokine which
inhibits HIV-1 replication in CD4 T cells and macrophages, like
IFN- (4). IFN- and IL-27 preferentially inhibit HIV replication in macrophages compared to CD4 T cells (5). Since the
mechanism of antiviral effect by IFN- is well investigated (69),
in order to understand the mechanism by which IL-27 inhibits
HIV-1 in both cell types, we compared the gene expression profiles
between IFN-- and IL-27-treated cells using a bioinformatics
analysis tool, DAVID (Database for Annotation, Visualization,
and Integrated Discovery) (5, 10).
DAVID (http://david.abcc.ncifcrf.gov) is an online application,
consisting of an integrated biological knowledge base and several
analytic modules aimed at systematically extracting enriched
biology from large gene/protein lists. Notably, DAVID is able to
handle large gene lists derived from any genomics platform (i.e.,
promoter microarray, proteomic data, and ChIP-on-CHIPs), and
therefore is not necessarily limited to the microarray platform
alone. The key foundation of DAVID enrichment analysis is that if
a biological process is disturbed in a given study, the co-functioning
genes should have a higher chance to be selected as a functional
group by the high-throughput screening technologies. This rationale
can make the analysis of large gene lists move from an individual
gene-oriented analysis to a gene functional group-based analysis.
Since the analytic conclusion is based on a group of co-functioning
genes instead of on an individual gene, it increases the likelihood
for investigators to identify the correct biological processes most
relevant to the study among the potential noise associated with
high-throughput experiments. More details regarding the data
analysis algorithms and backend databases implemented in DAVID
can be found in its original research publication (10), as well as in
the review articles (11, 12).
The false-positive results of microarray studies can be minimized
if attention is focused on avoiding as much random noise as
possible by controlling the sample processing environment and by
batching samples in such a way as to help neutralize the impact of
random gene selection. Thus, microarray study design should be
given careful attention and is dependent on an investigators goal
and funding. There are three main design approaches, which are
not mutually exclusive, that an investigator may choose when starting a microarray study: (1) a pilot study, (2) hypothesis generating
study, and (3) hypothesis testing study (see Note 1). A pilot study
was used to characterize the biological effects induced by IL-27 in
Interleukin-27 Induces Interferon-Inducible Genes
27
cultured primary CD4 T cells and macrophages as discussed herein,

and we have shown that IL-27 induces multiple IFN-inducible
genes (IFIGs) in macrophages and suggest that IL-27 inhibits HIV
replication in macrophages via a mechanism similar to that of
IFN- (5).
2. Materials
2.1. Isolation of CD4 T
Cells and Monocytes,
and Activation of
CD4+ T Cells and
Differentiation of
CD14+ Monocytes into
Macrophages
1. Lymphocytes separation medium (LSM, MP Biomedicals

LLC).
2. Phosphate-buffered saline (PBS), pH 7.2.
3. MACS buffer: sterilized PBS supplemented with 0.5% BSA
(Sigma-Aldrich) and 2 mM EDTA. Keep the buffer cold
(48C).
4. Human CD4 Microbeads (Miltenyi Biotec), Human CD14
Microbeads (Miltenyi Biotec), MACS LS columns (Miltenyi
Biotec), MACS Separator (Miltenyi Biotec).
5. Trypan blue solution (Invitrogen).
6. Complete RPMI-1640 (RP-10): RPMI-1640 (Invitrogen)
supplemented with 10 mM HEPES (Quality Biological Inc),
10% (v/v) of fetal bovine serum (FBS) (Hyclone) and 50 g/
mL of gentamycin (Invitrogen).
7. Complete D-MEM (D-10): D-MEM (Invitrogen) supplemented
with 10 mM HEPES (Quality Biological Inc), 10% (v/v) of FBS
(Hyclone) and 50 g/mL of gentamycin (Invitrogen).
8. Versen (Invitrogen), keep the buffer cold (48C).
9. PHA (Sigma-Aldrich, 5 g/mL in PBS) (100 Stock solution).
10. Recombinant human IL-2 (Roche Diagnostic) (10,000 units/mL).
11. Human AB serum (Grenfell, Gemini Byproduct).
2.2. Stimulation
of Cells with Cytokines
and Preparation of Cell
Lysate
1. Recombinant IL-27 (R&D systems) reconstituted at 20 g/mL

in PBS containing 0.1% BSA (Sigma-Aldrich).
2. Recombinant IFN- (IFN-2b from R&D systems) at 1 106
units/mL.
3. Warm RP-10 (37C).
4. Warm D-10 (37C).
5. Cold PBS (48C).
6. RLT buffer from QiAmp RNA mini kit (Qiagen) supplemented
with 1% (v/v) 2-mercaptoethanol (2-ME).
2.3. Microarray Data

Analysis
1. A computer (PC recommended) with 2 GB RAM, at least

2.0 GHz CPU, 10 GB free disk space and high-speed internet
access.
28
T. Imamichi et al.
2. Assisting software packages: MS Excel, GCOS, Affymetrix

Power Tool, MS Internet Explorer, Partek Pro., etc.
3. Methods
To avoid getting false-positive results, cells from independent
donors were treated with each cytokine. In a pilot experiment (see
Note 1), two experiments were performed. In a hypothesis
Generating Study (see Note 1), a total of six independent experiments were carried out.
3.1. Isolation
of Peripheral Blood
Mononuclear Cells
Peripheral blood mononuclear cells (PBMCs) are isolated from the

lymphocyte apheresis leukopacks of HIV uninfected healthy donors
from a blood bank using LSM. Cell viability and cell number are
determined by a trypan blue exclusion test.
1. In a biological safety cabinet, 30 mL of leukopack preparation
is carefully over-laid onto 10 mL of LSM in a 50 mL centrifuge
tube (Falcon).
2. Tubes are centrifuged at 400 g for 30 min at room temperature without brake.
3. Aspirate the top layer of clear plasma to within 23 mm above
the lymphocyte layer.
4. Transfer the lymphocyte layer (PBMCs) plus half of the LSM
layer below it to a centrifuge tube.
5. Add an equal volume of PBS to the lymphocyte layer in the
centrifuge tube.
6. Centrifuge for 10 min at 300 g at 4C.
7. Wash the cells again with PBS and resuspend in 50 mL PBS.
8. Take 10 L of PBMCs suspension, and mix with 10 L of trypan
blue solution. Allow dilution to incubate 23 min at room
temperature.
9. Apply 10 L of the mixture to a hemocytometer. Ensure that
the hemocytometer is not overfilled.
10. Remove the hemocytometer from the cabinet and place it on
the stage of a binocular microscope and focus on the cells.
11. Count the unstained cells (viable cells) and stained cells (dead
cells) separately in the hemocytometer. Count to a total at least
to 100 cells.
12. Calculate cell density.
3.2. Isolation of CD4+

T Cells and Activation
of the T Cells
In general, 20% of total PBMCs are CD4 T cells, thus if 100 106
CD4 T cells are required, a total of 500 106 PBMCs are
processed.
29
1. Determine the cell number and centrifuge the cell suspension

at 300 g for 10 min at 4C.
2. Aspirate PBS from the tube and then resuspend cell pellet in
80 L of cold MACS buffer per 107 total cells.
3. Resuspend the cells using a P-1000 pipette. Ensure to make
cell suspension without vortexing.
4. Add 20 L of CD4 MicroBeads per 107 cells.
5. Mix well and incubate for 15 min at 48C.
6. Wash cells by adding 2 mL of MACS buffer per 107 cells, and
centrifuge at 300 g for 10 min.
7. Aspirate the supernatants and then resuspend up to 108 cells in
500 L of MACS buffer using a P-1000 pipette.
8. Place a LS column in the magnetic field of MACS separator.
9. Equilibrate column by washing with 3 mL of MACS buffer.
10. Apply cell suspension onto the column.
11. Discard unlabeled cells which pass through and wash column
three times with 3 mL of MACS buffer. Perform washing step
by adding buffer three times, ensuring each time that the column
reservoir is empty.
12. Remove the column from the separator and place it on a 15 mL
centrifuge tube.
13. Add 5 mL of MACS buffer on the column, and then immediately flush out fraction with the magnetically labeled cells by
firmly applying the plunger supplied with the column.
14. Wash the eluted cells with 10 mL of warm RP-10 at 500 g for
5 min at room temperature.
15. Count cell number using trypan blue exclusion test (see step 8
in Subheading 3.1).
16. Make cell density at 2 106 cells/mL in T75 flask and add a
final 5 g/mL of PHA.
17. Culture the cells at 37C for 3 days in 5% CO2.
3.3. Isolation of CD14+
Monocytes and
Differentiated into
Macrophages
In general, 30~40% of total PBMCs are CD14+ monocytes, thus if

100 106 monocytes are required, a total of 200~300 106 PBMCs
are processed.
1. Determine the cell number and centrifuge the cell suspension
at 300 g for 10 min at 4C.
2. Aspirate PBS from the tube and then resuspend cell pellet in
80 L of cold MACS buffer per 107 cell total cells.
3. Resuspend the cells using a P-1000 pipette. Ensure to make
cell suspension without vortexing.
4. Add 20 L of CD14 MicroBeads per 107 cells.
30
T. Imamichi et al.
5. Mix well and incubate for 15 min at 48C.

6. Wash cells by adding 2 mL of MACS buffer per 107 cells, and
centrifuge at 300 g for 10 min.
7. Aspirate the supernatants and then resuspend up to 108 cells in
1,000 L of MACS buffer using a P-1000 pipette.
8. Place two LS columns in the magnetic field of MACS separator.
9. Equilibrate column by washing with 3 mL of MACS buffer.
10. Apply 0.5 mL of cell suspension per the column.
11. Discard unlabeled cells which pass through and wash column
with 3 3 mL of MACS buffer. Perform washing step by
adding buffer three times, ensuring each time the column
reservoir is empty.
12. Remove the column from the separator and place it on a 15 mL
centrifuge tube.
13. Add 5 mL of MACS buffer on the column, and then immediately flush out fraction with the magnetically labeled cells by
firmly applying the plunger supplied with the column.
14. Wash the eluted cells with 10 mL of warm D-10 at 500 g for
5 min at room temperature.
15. Count cell number using trypan blue exclusion test (see step 8
in Subheading 3.1).
16. Make cell density 0.8 106 cells/mL in warm D-10 with 10%
(v/v) of Human AB serum.
17. Transfer 10 mL of the cell suspension per Petri dish and incubate for 710 days at 37C in 5% CO2.
18. Aspirate culture supernatants and wash the cell monolayer
three times with 10 mL of PBS.
19. Add 5 mL of the cold Versen into the plates and then incubate
them at 4C for at least 1 h.
20. Using scrapers, remove the attached cells from plates.
21. Transfer the cell suspension to centrifuge tubes and wash cells
with warm D-10 at 500 g for 5 min.
22. Determine the cell density using trypan blue exclusion test
(see step 8 in Subheading 3.1).
23. Resuspend the macrophage fractions at 0.75 106 cells/mL in
D-10.
24. Add 2 mL per well in six-well plate and then incubate cells over
night at 37C in 5% CO2.
3.4. Stimulation
of CD4 T Cells with
IL-27 or IFN-a
1. 3 days after incubation, the PHA-stimulated CD4 T cells are

collected from the flasks.
2. The cells are washed with warm RP-10 at 500 g for 5 min.
31
3. Resuspend cell pellet and determine cell density and viability.

4. Adjust cell density to 4 106 cells/mL in RP-10.
5. Put 2.5 mL of cell suspension in a 50 mL conical tube and then
culture for 46 h in RP-10 to make cells rest.
6. During incubation, the following solutions are made and
prewarm at 37C until use: media, IL-27 (200 ng/mL) or
2,000 units/mL of IFN-.
7. After preincubation, add 2.5 mL of media or cytokine solution
to tubes containing cells.
8. Vortex cells to homogenize and then culture cells for 24 h.
9. After incubation, 10 mL of cold PBS is added in each tube followed by vortexing to stop cell reaction.
10. Cells are washed with 10 mL of cold PBS three times, and then
cell pellets are resuspended in 1 mL of cold PBS and transferred into 1.5 mL RNaseDNase-free tubes.
11. Spin the tubes at 500 g for 5 min at 4C and then remove all
liquid from the tube.
12. Add 0.35 mL of RLT buffer containing 1% (v/v) of 2-ME and
lysis cells using a P-1000 pipette, and store them at 80C.
3.5. Stimulation of
Monocyte-Derived
Macrophages with
IL-27 or IFN-a
1. Seven to 10 days after incubation, nonadherent cells are aspirated

and then the attached cells are washed using 10 mL of room
temperature PBS, three times.
2. After soaking the last wash with PBS, add 5 mL of the cold
Versen per plate and then incubate the plate for 1 h at 4C.
3. After the 1-h incubation, adherent cells are harvested using cell
scrapers.
4. Cell suspensions are collected in a 50 mL conical tube and
washed with warm D-10 at 500 g for 5 min at room
temperature.
5. Repeat step 4 and then determine cell density.
6. 1.5 106 MDMs per well are seeded onto a six-well plate, and
then cultured over night at 37C in DMEM-10.
7. On the next day, wash each well with warm D-10 and then add
1.5 mL of warm media alone (for control), 1,000 units/mL of
IFN-, or 100 ng/mL of IL-27.
8. Cells were cultured for 24 h in the incubator.
9. After incubation, 2 mL of cold PBS is added per well to stop
cell reaction, and sack all liquids.
10. Wash wells with 3 mL of cold PBS and sack the liquid.
11. Add 0.35 mL of RLT buffer containing 1% (v/v) of 2-ME per
well.
32
T. Imamichi et al.
12. Make complete cell lysate using a scraper and transfer all cell
lysate into 1.5 mL RNaseDNase-free tubes, and then store
the cell lysate at 80C until processing RNA extraction.
3.6. RNA Extraction
from the Frozen Cell
Lysate
1. Thaw the cell lysate in RLT at room temperature.

2. Add 1 volume of 70% (v/v) ethanol to the homogenized lysate,
and mix by pipeting. Do not centrifuge (see Note 2).
3. Apply up to 700 L of the sample, including any precipitates
that may have formed, to an RNeasy Mini column placed in a
2 mL collection tube provided by the kit. Close the tube gently,
and centrifuge for 15 s at >8,000 g at room temperature.
Discard the flow-through.
4. Transfer the spin column into a new 2 mL collection tube, and
add 700 L of buffer RW1 to the column. Close the tube
gently, and then centrifuge for 15 s at >8,000 g to wash the
column. Discard the flow-through and the collection tube.
5. Transfer the column into a new 2 mL collection tube. Pipet
500 L of Buffer RE onto the RNeasy column. Close the tube
gently, and centrifuge for 15 s at >8,000 g to wash the column.
Discard the flow-through.
6. Add another 500 L of Buffer RE to the RNeasy column.
Close the tube gently, and then centrifuge for 8 g to dry the
column.
7. Transfer the RNeasy column to a new 1 mL collection tube,
and then centrifuge at full speed for 1 min to eliminate any
chance of possible Buffer RE carryover.
8. To elute RNA from the column, transfer the column to an
RNase-free 1.5 mL tube. Pipet 50 L of RNase-free water
directly onto the column. Close the tube gently, and centrifuge
for 1 min at >8,000 g. RNA amounts are quantitated by OD
reading.
3.7. RNA Expression

Analysis Using
Affymetrix Gene Chip
The isolated RNA was labeled and hybridized on Affymetrix

Hu133 plus 2.0 arrays following the Affymetrix one cycle target
labeling protocol in the expression manual (https://www.affymetrix.com/support/downloads/manuals/expression_analysis_
technical_manual.pdf).
3.7.1. Affymetrix One Cycle

Target Labeling
(a) First strand cDNA synthesis.

1. Place 5 g of the total RNA in a PCR tube.
2. Add 2 L of the appropriately diluted poly-A RNA. (The
eukaryotic Poly-A RNA spike-in control could provide the
exogenous control for monitoring the eukaryotic target
labeling process. The table below shows the serial dilution
according to the amount of starting total RNA).
33
Serial dilutions
Starting amount
total RNA (mg)
First
Second
Third
Spike-in
amount (mL)
1:20
1:50
1:50
1:20
1:50
1:10
10
1:20
1:50
1:5
3. Add 2 L of 50 T7-Oligo(dT) Primer.

4. Add RNase-free H2O to a final volume of 12 L.
5. Flick tube a few times to mix well and centrifuge briefly.
6. Incubate the mixture for 10 min at 70C and cool to 4C
for 2 min.
7. Transfer 7 L of First Strand mix (4 L of 5 first Strand
Reaction Mix; 2 L of 0.1 M DTT; and 1 L of 10 mM
dNTP) into the sample RNA/T7 mixture and flick to mix
and centrifuge briefly.
8. Incubate for 2 min at 42C.
9. Add 1 L of Superscript II (if the starting total RNA is
18 g) or 2 L of Superscript II (if the starting total RNA
is 8.115 g) into each tube.
10. Incubate for 1 h at 42C and cool to 4C for 2 min.
11. Centrifuge briefly to collect samples at the bottom.
(b) Second strand cDNA synthesis.
1. Assemble Second-Strand Master Mix as shown below per
sample. The second strand master mix is recommended to
use immediately after making. Add additional material to
compensate for loss during the process when making
master mix.
Reagent
Volume (mL)
RNase-free H2O
91
5 second Strand Reaction Mix
30
dNTP (10 mM)
E. coli DNA Ligase
E. coli DNA Polymerase I
RNase H
Total volume
130
2. Transfer 130 L of Second-Strand Master Mix to each

first-strand synthesis sample (total volume 150 L).
3. Incubate for 2 h at 16C.
4. Add 2 L of T4 DNA Polymerase to each reaction.
34
T. Imamichi et al.

6. Add 10 L of EDTA (0.5 M) and mix well to terminate
reaction.
(c) Cleanup double-stranded cDNA.
1. Transfer the 150 L ds cDNA reaction to a new 1.5 mL
tube for each sample.
2. Add 600 L of cDNA Binding Buffer to the cDNA synthesis reaction. Mix by vortexing for about 3 s.
3. Add 500 L of the sample to the cDNA Spin column and
centrifuge 1 min at 8,000 g. Discard flow-through.
4. Add remaining mixture onto the same spin column and
centrifuge 1 min at 8,000 g. Discard flow-through.
5. Add 750 L of cDNA Wash Buffer (ensure ethanol has
been added to Wash Buffer) to the spin column and centrifuge 1 min at 8,000 g. Discard flow-through.
6. Open the cap of the spin column and centrifuge (place
columns in every second well of rotor) for 5 min at max
speed (<25,000 g) to dry the columns. Discard the flowthrough and collection tube.
7. Transfer spin column into a new 1.5 mL collection tube
and add 14 L of cDNA Elution Buffer directly onto the
center of the column membrane.
8. Incubate 1 min at room temperature and centrifuge 1 min
at max speed.
9. The collection of the ds cDNA was used for next step:
Synthesis of Biotin-Labeled cRNA.
(d) Biotin labeling antisense cRNA.
1. Determine the amount of cDNA to be used for each IVT
reaction from the table below.
Starting total RNA (mg)
Volume of cDNA to use in IVT
1.08.0
All (~12 L)
8.115
6 L
2. Transfer the needed amount of template to microcentrifuge

tube and add the IVT synthesis reaction components in
the order indicated in the table below. Do NOT prepare
on ice.
Reagent
Volume (mL)
Template cDNA
Calculated from step 1

above
RNase-free H2O
Variable (to give volume

of 40 L)
(continued)
Reagent
Volume (mL)
10 IVT Labeling Buffer
IVT Labeling NTP Mix
12
IVT Enzyme Mix
Total volume
40
35
3. Carefully mix the reagents and briefly centrifuge.

4. Incubate at 37C for 16 h. Labeled cRNA can be stored
at 20C overnight or 70C for long term.
(e) Cleanup biotinylated cRNA.
1. Add 60 L of RNase-free H2O to the IVT reaction and
mix by vortexing for about 3 s.
2. Transfer sample plus water to a 1.5 mL tube.
3. Add 350 L IVT cRNA Binding Buffer and mix by
vortexing.
4. Add 250 L ethanol (100%) and mix by pipetting.
5. Add sample (~700 L) to IVT cRNA Cleanup Spin Column
and centrifuge 15 s at 8,000 g, discard flow-through.
6. Add 500 L of IVT Wash Buffer (ensure ethanol has been
added to Wash Buffer) to the spin column and centrifuge
1 min at 8,000 g. Discard flow-through.
7. Add 500 L of 80% ethanol onto the spin column and
centrifuge 15 s at 8,000 g. Discard flow-through.
8. Open the cap of spin column and centrifuge (place columns
in every other well) for 5 min at max speed (<25,000 g)
to dry the columns.
9. Transfer spin column into a new 1.5 mL collection tube
and add 11 L RNase-free H2O directly onto the center of
the column membrane.
10. Centrifuge 1 min at max speed to elute.
11. Add 10 L of RNase-free H2O directly onto the spin
column membrane.
12. Centrifuge 1 min at max speed to elute. The cRNA could
be stored in 20C or 70C if not used immediately.
13. The quality and quantity of the labeled cRNA could be
measured on Bioanalyzer and Nanodrop, respectively.
3.7.2. Hybridization on
Affymetrix Array
(a) cRNA fragmentation.

1. Assemble Fragmentation Buffer as shown in the table
below.
36
T. Imamichi et al.
Reagent
Volume
cRNA
1520 g (121 L)
5 Fragmentation Buffer
RNase-free H2O
Variable to 40 L final vol.
Total Volume for 1 rxn
40
2. Incubate at 94C for 35 min.

3. Save 2 L of the fragmentation sample to check on
Bioanalyzer.
4. Fragmented cRNA could be stored at 20C or 70C.
(b) Hybridization.
1. Equilibrate the probe array to room temperature before
using. Prewet the probe arrays with prehybridization mix
and put in hybridization oven.
2. Preparing Target Hybridization Cocktail as shown in the
table below.
Reagent
Volume
Fragmented and Labeled cRNA
15 g
Control Oligonucleotide B2 (3 nM)
5 L
20 Eukaryotic Hybridization
Controls (bioB, bioC, bioD, cre)
15 L
2 Hybridization Mix
150 L
DMSO
30 L
Nuclease-free Water
To final volume
300 L
Total volume
300 L
The frozen 20 hybridization controls were heated at

65C for 5 min before using.
3. Heat hybridization sample at 99C for 5 min.
4. Incubate the sample at 45C for 5 min.
5. Spin samples at max RPM for 5 min at room temperature
to remove insoluble material from hybridization cocktail.
6. Remove buffer solution from probe array cartridge and fill
with 200 L of hybridization cocktail.
7. Transfer the arrays into hybridization oven to hybridize at
45C for 1618 h at 60 rpm (load the probe arrays in a
balanced configuration in the oven).
3.8. General Guidelines
of Microarray Data
Analysis
Through over a decade of experience and the running of over

13,000 Affymetrix microarrays, we view each discovery-oriented
37
microarray study as an individual dynamic procedure, each with its

own personality that needs to be understood in order to extract the
most meaningful information. The data analysis and functional
annotation approach to a microarray study is more of an art than it
is a set of concrete statistical procedures and workflows. For each
study, we will often look at a dozen or so preprocessing methods
and statistical thresholds, each of which is followed up with functional annotation analysis. The combination of results from such an
analysis taken together with anticipated biological changes, an
investigators general background knowledge, and old-fashion scientific intuition allows one to settle into a final gene selection criterion that yields a good compromise between controlling the
false-positive rate and the loss of true positives, i.e., a good balance
between Type I and II Errors. The scientific reasoning for taking
this approach is several fold in that each study has its own level of
(1) systematic noise that is influenced by sample handling, laboratory and clinical setting, reagent lots, technical experience, tissuespecific expression complexity and background hybridization, etc.;
(2) gene-to-gene interdependencies (one gene can cause multiple
genes to increase and others to decrease, with each of those genes
influencing expression levels of additional genes, and so on); (3)
genes that have normal versus non-normal intensity distributions
which impacts the number of genes passing a specific statistical p
value threshold; (4) probes that bind to their target at specific
hybridization efficiencies since a single hybridization condition is
used for the binding of hundreds of thousands of probe-target pairs.
Thus, there is not a single set of preprocessing methods and statistical thresholds that will work for every study (see Table 1 for General
Table 1
A guideline for determining the number of samples
Absolute
log2 fold
change
Study type
Cell type
Number of
samples % Present P value
Pilot
Treated cell lines

Treated primary cells
In vivo isolated cells
12
24
48
100
>75
>60
NA
<0.05
<0.05
>1.02.0
>1.0
>0.6
Hypothesis
Treated cell lines
generating Treated primary cells
35
510
>10
>75
>60
>50
<0.01
<0.05
<0.05
>1.02.0
>0.6
>0.4
Hypothesis
testing
>10
>20
>30
>90
>90
>90
(p value x # of genes) <0.05 >0.6

Treated cell lines

Treated primary cells
General guideline for determining the number of samples to use for each condition under study and for
statistical thresholds that designing and analyzing a microarray study. % Present refers to the percent of
samples for which a given gene is called Present by an analysis package such as Affymetrixs GCOS and
Gene Expression Console
38
T. Imamichi et al.
Guidelines for selecting statistical thresholds). The caveat to such a

post hoc data mining approach is that the results should be followed up with confirmation studies. Such a mindset has greatly
decreased our false-positive confirmation rate and has more rapidly
led to new findings and publications than if we had used a standard
of methods and statistics that were set a prior to the data analysis.
3.9. Gene Selection
To compare the gene expression profile between IL-27 and IFN-

treated CD4 T cells and macrophages, as mentioned above, a pilot
study was initiated. We performed two experiments on CD4 T cells
and macrophages treated with 100 ng/mL of IL-27 or 1,000 units/
mL of IFN-, respectively. The samples were hybridized on
Affymetrix Human genome U133plus 2.0 arrays (see Note 3).
Data were collected in the Affymetrix GCOS system after scanning
the array. Data analysis followed the steps as below:
1. The normalization and background correction were done in
the Affymetrix GCOS system using the standard MAS5
algorithm (14).
2. Compared IL-27 or IFN- treatment samples in the same cell type
as its mock control sample, such as IL-27 treated vs. control and
IFN- treated vs. control within both CD4 cells and macrophages.
3. The selection focused on a number of Affymetrix calls. They are
(a) Signal; as the intensity for the expression of each gene. (b)
Detection Call; indicating if the transcript was detected (A
representing Absent, M representing Marginal, and P representing Present). (c) Change; indicating up or down regulation in the comparison of two samples (D representing
Decrease, MD representing Marginal Decrease, NC representing No Change, MI representing Marginal Increase, and
I representing Increase). (d) Signal Log Ratio; the log2 fold
change indicating up or down regulation in the comparison of
two samples. In this pilot experiment, after comparing the treatment sample to its control within the same cell type, the selection criteria are (1) detection call is not A, (2) change is not
NC, and (3) signal log ratio is larger than 1 or less than 1.
4. There were 132 transcripts up-regulated and 78 transcripts
down-regulated by IFN-, 37 transcripts up-regulated and
137 transcripts down-regulated by IL-27 in CD4 cells, 438
transcripts up-regulated and 683 transcripts down-regulated
by IFN-, and 662 transcripts up-regulated and 935 transcripts
down-regulated by IL-27 in macrophages (a gene list is available at http://david.abcc.ncifcrf.gov/manuscripts/cp/).
5. All selected transcripts above were classified by the Hierarchical
clustering using Partek Pro statistical software (www.partek.
com) (Fig. 1).
6. The common genes among the lists derived from step 4 can be
compared by Venn diagram (Fig. 2).
39
Fig. 1. Heap map from microarray analysis. CD4 T cells or macrophages were treated with
mock, 100 ng/mL IL-27, or 1,000 units/mL IFN- for 24 h. Differentially expressed genes
were selected between mock and cytokine-treated cells and were greater than a twofold
change. The resulting retention of 1,868 genes out of approximately 54,000 AffyID was
subjected to subsequent analysis. An increase or decrease of gene transcription by more
than twofold is represented in red or green, respectively. Genes shown in black indicate
no change in transcriptional activity.
Fig. 2. Venn Diagram of upregulated genes between IFN- and IL-27 treated macrophages.
MDM cells were treated with IFN- and IL-27, respectively. Approximately 438 and 662
up-regulated genes were selected according to the microarray experiments.
3.10. Gene Functional

Annotation with DAVID
In contrast to the traditional approach of studying one or a few

genes at a time, the advantage of the high-throughput genomic
microarray technology is to allow investigators to simultaneously
measure the changes and regulation of genome-wide genes under
certain biological conditions. After interesting gene lists are
obtained from microarray experiments, DAVID systematically maps
a large number of genes in a given list to the associated biological
annotation terms (e.g., GO Terms or Pathways), and then iteratively
examines each of the annotation terms by the Fisher Exact test.
Thereafter, the annotation terms with enriched gene members can
be identified from tens of thousands of other annotation terms in a
high-throughput manner. The enriched annotation terms associated with the given gene list will give important insights for investigators to understand the biological themes under the study (see
Note 4 for a hypothetical example of an enrichment analysis).
40
T. Imamichi et al.
Importantly, given that the high-throughput enrichment

data-mining environment is extremely complicated (12), the
analysis of large gene lists is indeed more of an exploratory procedure rather than a purely statistical solution. The analysts themselves still play critical roles in making the final decisions in terms
of which enriched biology (terms) make more sense to a given
study, and thereafter which enriched terms to follow up and focus
on. Even though annotation terms may be associated with very significant enrichment p values, it is common that analysts ignore
some of the enriched annotation terms based on whether or not
the results make sense, biologically. The analogous example of this
type of situation is like that of a Google search, which returns some
results that are not relevant to the users original query. Users
therefore can make the final judgment to ignore some of the results
based on his or her knowledge of the situation.
Given that high-throughput gene functional annotation analysis
is an exploratory procedure, rather than a strictly defined protocol,
the following section describes major steps and procedures using
DAVID in order for readers to grasp the idea of the overall data
analysis environment that is available, as well as the key spirit of
high-throughput gene functional annotation analysis. As for the
finer level of DAVID functions that are available, they may be
explored in a logical manner throughout the course of the analysis
by the reader.
1. Get gene list(s) ready for functional analysis with DAVID.
A typical gene list is usually in a size ranging from hundreds to
thousands of genes, which correspond to certain biological
themes under study (see Note 5 for details of IFN_Up_List
and IL27_Up_List which will be used throughout descriptions
and discussions below).
2. Submit list(s) of gene IDs to DAVID at http://david.abcc.
ncifcrf.gov or http://david.niaid.nih.gov. After clicking on
Start Analysis on the header, a gene list manager panel will
appear on the left side of the web page (Fig. 3). Then, perform
the following steps to submit a gene list to the DAVID system:
Step 1. Copy and paste a list of gene IDs into box A (i.e.,
Affymetrx_ID). Step 2. Indicate the list to be submitted as a
gene list (i.e., genes to be analyzed). Step 3. Click the Submit
List button (see Note 6).
3. Use the Tool Menu Page as the central page to access various
DAVID analytic tools (Fig. 4) (see Note 7).
4. Invoke Gene Name Batch Viewer to explore the names of all
genes, particularly for highly expected and related genes in the
list (see Note 8).
5. Invoke Gene Functional Classification to classify individual
genes into gene functional groups based on the overlap of
41
Fig. 3. The gene list submission page of DAVID. The right side of panel lists a format example of Affymetrix gene list. The
left side is the gene list manager. To submit a gene list to DAVID, the gene list can be copied/pasted to the box followed by
steps 1, 2, 3, and 4 as labeled.
similar biological function. Users can treat this step as a clustering view of step 4, that is, highly functionally related genes
are grouped together for the ease of exploration. To do this,
return to the Tool Menu Page as described in step 3. Click on
Gene Functional Classification Tool to classify the input
gene list into gene groups (see Note 9 and Fig. 5).
6. Invoke Functional Annotation Chart to understand the
fine details of enriched annotation terms associated with
the large gene list (see Note 10 and Fig. 6).
42
T. Imamichi et al.
Fig. 4. The tool menu page of DAVID. After the gene list(s) are submitted to DAVID, the gene list manager on the left side
displays appropriate information. The hyperlinks of four major DAVID tools (pointed by arrow icons) are listed on the right
side. The strengths and indications of DAVID tools are various for different analytic goals as discussed in original papers
(10), as well as in this protocol. By clinking on the hyperlink of DAVID tools, users can invoke the according DAVID tool to
analyze the current gene list that is being highlighted (in blue) in the left gene list manager. Importantly, by clicking on
Start Analysis on the menu, users can get to this page to access/switch DAVID tools at any time during analytic course.
7. Invoke Functional Annotation Clustering (Fig. 7) to explore

enriched annotation terms in a clustered view instead of a
linear term view as in step 6 (see Note 11).
8. Compare the annotation profiles across relevant gene lists
(see Note 12, Table 2 and Fig. 9).
43
Fig. 5. The layout of the result page of DAVID Gene Functional Classification Tool. The 443 genes in IFN_Up_List were classified
into multiple gene functional classes that are separated by the blue rows. The genes in each functional class should share
significant amount of biological functions (terms). Variety of hyperlinks is provided for each functional class and its gene
members. Such clustering view make genes in users list much organized for the ease of exploration and focus.
44
T. Imamichi et al.
Fig. 6. The annotation summary page. According to research interests, the wide range of annotation categories, offered by
DAVID, can be selected/deselected through the expandable tree structure by clicking on + icons. Then, three analytic
modules (three buttons on the bottom) can be respectively invoked to analyze gene list (e.g., IFN_Up_List) against the
selected annotations above.
Table 2
Comparison of the top enriched terms for IFN-a and IL-27 gene lists
Top enriched biology from DAVID chart report
IFN_Up_List
IL27_list
Data source
Annotation term
SP_PIR_KEYWORDS
Interferon induction
GOTERM_BP_3
p value
p value
7.57
1.91E-43
6.26
7.37E-43
Defense response
23.24
8.86E-29
24.19
2.35E-39
GOTERM_BP_3
Immune response
22.16
9.02E-29
22.68
7.32E-38
GOTERM_BP_3
Response to pest, pathogen,

or parasite
13.78
1.70E-21
13.39
2.04E-25
GOTERM_BP_3
Response to other organism
14.05
3.98E-21
13.39
5.32E-24
(continued)
45
Table 2
(continued)
Top enriched biology from DAVID chart report
IFN_Up_List
IL27_list
Data source
Annotation term
p value
p value
SP_PIR_KEYWORDS
Alternative splicing
25.95
2.03E-10
27.65
1.45E-15
SP_PIR_KEYWORDS
Direct protein sequencing
16.22
4.56E-08
16.85
8.13E-11
SP_PIR_KEYWORDS
Membrane
22.16
1.42E-05
25.27
7.52E-11
SP_PIR_KEYWORDS
Signal
16.22
2.39E-05
17.93
8.78E-09
SP_PIR_KEYWORDS
Metal-binding
14.59
9.34E-05
NA
NA
GOTERM_BP_3
Response to wounding
6.49
7.30E-07
SP_PIR_KEYWORDS
Zinc
12.16
0.000173
SP_PIR_KEYWORDS
Chelation
1.08
0.000184
1.08
9.21E-06
SP_PIR_KEYWORDS
Transmembrane
20.54
0.000215
23.33
1.77E-08
SP_PIR_KEYWORDS
Hydrolase
10.81
8.65E-06
9.94
1.93E-05
SP_PIR_KEYWORDS
Innate immunity
1.62
0.000255
2.16
3.26E-08
GOTERM_BP_3
Positive regulation of cellular

process
6.76
3.67E-05
6.05
7.85E-05
SP_PIR_KEYWORDS
Transmembrane protein
6.49
0.000295
7.34
1.02E-06
SP_PIR_KEYWORDS
Immune response
2.70
0.000531
4.75
1.60E-12
SP_PIR_KEYWORDS
Antiviral defense
0.81
0.003541
1.94
1.17E-07
SP_PIR_KEYWORDS
Glycoprotein
17.57
0.003609
22.03
8.08E-09
GOTERM_BP_3
Positive regulation of physiological process
5.95
0.000119
4.75
0.002229
SP_PIR_KEYWORDS
Surface antigen
1.08
0.020628
1.73
7.95E-06
SP_PIR_KEYWORDS
sh2 Domain
1.35
0.032123
2.16
2.04E-05
SP_PIR_KEYWORDS
ubl Conjugation
2.70
0.000215
1.73
0.014942
GOTERM_BP_3
Hemopoietic or lymphoid
organ development
2.43
0.000221
1.51
0.016281
5.83
NA
9.77E-07
NA
The top 20 enriched terms from each of DAVID Chart Reports for IFN_Up_List and IL27_Up_List are selected and
combined. The according enrichment p values and gene hit percentages are listed side-by-side. There are large agreements
on the very relevant immune-related terms between the two lists. Two terms of metal-binding and zinc are missing
from IL-27 study. Considering the lists from immunology studies, the missing terms may not be very interesting. In such
case, analysts should make the final judgment based on overall situations instead of solely relying on statistical values
46
T. Imamichi et al.
4. Notes
1. Pilot Study. The goal of a pilot study can be several fold but is
most often performed in preparation for a larger study in order
to optimize workflow and experimental conditions so that time
and funds are most efficiently used to generate high-quality
data. Hypothesis Generating Study. This type of study focuses
on the discovery of genes and/or pathways not known to be
involved in the current biological phenomena under study with
the idea that the genes and pathways will lead to a new
hypothesis(es) that can be confirmed via follow-up laboratory
experiments. Hypothesis Testing Study. Hypothesis testing
studies are designed in such a manner as to have very tight
control over Type I Error, even at the cost of a high Type II
Error, i.e., attempt to eliminate any false-positive error even if
it means throwing out many true positives. Table 1 summarizes
the number of recommended samples and statistical thresholds
to use based on the scientific approach and sample type.
2. Visible precipitate may form after the addition of ethanol when
preparing RNA from certain cell lines, but this will not affect
the RNA extraction.
3. Affymetrix Hu133 plus 2.0 arrays contain from eleven to
twenty-one 25-mer oligonucleotide probes that are specific for
each gene being interrogated and are Perfect Match (PM)
probes. Additionally, a second set of probes identical to the
first, except for a single nucleotide change at the center position, is included to eliminate nonspecific binding signals and
are termed Mis-Matched (MM) probes. A group of such
probes specific for a given gene is called a probe set. There are
numerous published methods (such as MAS5, RMA, GCRMA,
etc. as reviewed by Harr et al. (13, 14)) available for Affymetrix
data preprocessing, probe set summarization, and statistical
analysis. There are several free software packages commonly
used for Affymetrix GeneChip preprocessing including
Affymetrixs Gene Expression Console and Affymetrix Power
Tools (for advanced users) which can be downloaded from the
Affymetrix website (www.affymetrix.com), and various R
statistic packages from Bioconductor (www.bioconductor.org).
4. In the previous gene statistical selection in Subheading 3.9,
gene expression was compared between untreated and
cytokine-treated cells. The gene selection statistical analysis
identified over a 1,000 genes regulated in one or both cytokine
treated MDM and CD4 T cells. Heat map analysis further
categorized them into multiple subgroups/gene clusters
with distinct up/down gene regulation patterns (Fig. 1).
47
For functional analysis of the gene lists, we now illustrate the

idea of DAVID enrichment analysis with a hypothetical
example. For example, 1,000 regulated genes are selected via
microarray study on IFN- treated cells, of which, 50 out of
1,000 genes (5%) are IFIGs. As compared to the global background of the microarray chip, on which there are 20,000
genes in total, of which, 100 out of the 20,000 genes (only
0.5%) are IFIGs. It is obvious that IFIGs are much more
strongly selected (5%) by the microarray experiment, than by
random chance (0.5%). The significance of the enrichment
p value can be mathematically measured by well-known statistical methods, such as the Fishers exact test (i.e., 2.5E-38).
A conclusion can then be obtained for the particular example,
that is, IFIGs are significantly enriched in the users gene list
and therefore should be relevant to the study.
5. The following example gene lists will be used throughout the
procedure. Approximately, 438 (IFN_Up_List) and 662
(IL27_Up_List) genes were identified as two subgene lists
from Subheading 3.9, as up-regulated genes for IFN- and
IL-27-treated MDMs, respectively. The genes in the two lists
are represented by Affymetrix probe set IDs. The list of gene
IDs should be in an acceptable format (i.e., comma delimitation,
space delimitation, or one ID per line). The detailed information and examples of supported common ID types and input
ID formats can be found on DAVID web site.
6. After the list of 438 genes has been successfully submitted to
DAVID, a gene list, named Uploaded_List_1, should appear in
the gene list manager panel. It can be manually renamed to a
more meaningful name such as IFN_Up_List as shown in
Fig. 4. Multiple gene lists can be uploaded to DAVID one after
another. Once gene lists are in the gene list manager, they can
be accessed by any of the DAVID tools. Users can also manually switch back and forth between gene lists during the course
of the analysis.
7. The Tool Menu Page provides a set of hyperlinks for four sets
of available analytic tools (Fig. 4). Clicking on each link will
lead to the corresponding DAVID tool for analysis of the
current working gene list (IFN_Up_List), which is highlighted in the gene list manager. Importantly, by clicking on
Start Analysis on the header menu, users can always go
back to this page for choosing or switching to other analytic
tools during the analysis course.
8. The purpose of this step is to give users a rough idea about the
contents of the gene list. This way, questions like Does my
gene list contain important marker genes expected for the
study? can be answered. To do this, click on the Gene Name
Batch Viewer link on the Tool Menu Page. All the gene names
48
T. Imamichi et al.
will be listed in a linear format. For a gene of interest, various

hyperlinks are provided for more detailed annotation information (e.g., other gene names, more gene symbols, other important
gene IDs, etc.). For IFN_Up_List, there are many immunology-related genes as expected for the study, such as chemokine
ligand genes, interleukin genes, and interferon inducible genes.
For example, the Affymetrix ID, 202086_at, corresponds to
the gene, myxovirus (influenza virus) resistance 1, interferoninducible protein p78 (mouse) (MxA). In addition, the hyperlink provided for the gene offers more summarized information
about MxAs IFN inducibility, antiviral ability, and OMIM
phenotype association, as well as other ID types and annotation
resources specific to this gene. After exploring through the
details of many interesting genes in the list as above, analysts
not only have a rough idea about the contents of the gene list,
but also know more details about particular genes of interest.
9. For IFN_Up_List, genes can be classified into multiple groups,
such as, an interferon-inducible gene group (group 2), and a
chemokine gene group (group 4). Various hyperlinks are also
provided for more detailed information for each gene/gene
group corresponding to users interests. Tools in step 4 and 5
in Subheading 3.10 give us a similar global view of gene contents in the IFN_Up_List. The advantage of Gene Functional
Classification is that it provides an organized clustering view of
input genes so that analysts can quickly grasp the key spirit of
the gene list by going through group-level information as
opposed to a gene-by-gene level, and thereafter easily focus on
genes of relevance. Analysts should keep in mind that it is not
recommended to replace step 4 with step 5 in Subheading 3.10
because some important genes may be left out from the clustering view if they do not have a strong network context with
other genes (Fig. 5).
10. This tool implements gene-term enrichment analysis as
described in the DAVID introduction section. This is a key
step among all others during the analytic course. To do this,
return to the Tool Menu paged as described in step 3 in
Subheading 3.10. Click on Functional Annotation Chart to
show the Summary Page. There are seven annotation categories (e.g., Go Terms, BioCarta Pathways, Proteinprotein
interactions, etc.), which will be subjected to enrichment analysis, and are preselected by default (Fig. 6). However, users
can always manually select and deselect any annotation categories through the expandable annotation category trees, according to their research interests. Next, click on the Functional
Annotation Chart button on the bottom of the page leading
to a Chart Report. For IFN_Up_List, the Chart Report lists
enriched annotation terms ordered by their enrichment p values.
49
Fig. 7. The layout of Chart Report. The enriched terms and their associated statistical values are listed in a linear tabular format.
They are ordered by the enrichment p values. The top ranked terms such as interferon induction, immune response, are
exactly what users expect for the study regarding IFN_Up_List. The hyperlinks on the terms lead to more detailed explanation
of the terms. Clicking on blue horizontal bars, the genes in users list that belong to the corresponding terms will be listed.
For example, the top five reported terms in 15 are interferon

induction, response to biotic stimulus, immune response, defense
response, and response to pest (Fig. 7). Given IFN_Up_List was
obtained from an IFN stimulation experiment; these terms are
exactly what we expected. Thereafter, analysts can focus on
particular annotation terms of interest and further ask questions like, What are the genes in my list that are associated
with this term? For example, by clicking on the blue bar
beside the term, interferon induction, ~28 genes associated
with the term can be explored (i.e., AIM2, CXCL10, DDX58,
EIF2AK2, G1P2, G1P3, GBP1, IFI16, IFI27, IFI35, IFI44,
IFIT1, IFIT2, IFIT3, IFIT5, IFITM1, IFITM2, IFITM3,
INDO, IRF1, ISGF3G,MX1, MX2, OAS1, OAS2, OAS3,
OASL, PSME1, RNF31, etc.). For another example, the genes
associated with the term, antiviral defense, include APOBEC3G,
DDX58, IFIH1, ISGF3G, RNF31, MX1, MX2, PLSCR1,
STAT1, and STAT2. Comparing the two sets of genes, analysts
may focus more on genes having both antiviral and IFN inducible characters. Moreover, the KEGG and BioCarta pathways
are other important sources that allow analysts to view their
genes in a network context. The enrichment p value for IFN
alpha signaling pathway is calculated and is found to be very
50
T. Imamichi et al.
Fig. 8. The genes in users list on biological pathway. For IFN_Up_List, BioCarta pathway
of IFN- Signaling Pathway is identified in the enrichment analysis with significant enrichment p value. The genes of STAT1, STAT2, p48 (indicated with red stars), regulated in this
study, are shown on the map. Thus, users can examine genes of interests in a biological
network context.
significant (i.e., 3.5E-5) in the Chart Report. A graphical representation of the pathway can be invoked for analysts to further explore regulated genes (e.g., STAT1 and p48) on the
pathway (Fig. 8). At this point, it is shown that STAT1 plays an
important role in IFN signal transduction in this study, which
is expected based on a priori knowledge. DAVID tools provide
an integrated and enriched data mining environment for analysts to identify the most relevant and important annotation
terms associated with large gene list(s) under study. Analysts
can extend the above example procedure and logic to many
other relevant and enriched annotation terms, in order to
obtain more comprehensive analytic results. Importantly, analysts themselves play the final decision-making role to judge
which terms are more interesting and relevant for further focus
based on a priori biological expectation and knowledge for the
given study.
11. The advantage of this function is to organize/cluster redundant
annotation terms with similar meaning (e.g., programmed cell
51
death, induction of apoptosis, apoptosis, and regulation of

apoptosis) into a clustered view for the ease of interpretation
and focus. To do this, go back to the Summary Page as
described in the beginning of step 6 in Subheading 3.10. Then
click on the Functional Annotation Clustering button on the
bottom of the page. Analysts can go through the important
enriched annotation terms in a clustered view in a similar manner as that of step 6 in Subheading 3.10. We will not repeat the
exploration procedure here again. In addition, as discussed in
step 5 in Subheading 3.10, the clustered view cannot totally
replace the linear view. It is recommended to explore both in
order to obtain maximum satisfactory results from the various
analyses.
12. The 662 up-regulated genes (IL27_Up_list) obtained from
IL-27 treated MDM in Subheading 3.9 can also be submitted
and explored through DAVID from step 1 to 7 as outlined for
IFN_Up_List. Then, the two sets of annotation results for
the IFN genes and IL-27 genes can be cross-compared.
This comparison is particularly interesting because it may
answer a key question asked by the study Does IL-27, a
novel anti-HIV cytokine, share similar mechanisms with
IFN-? For other studies, with a time series design, such
comparisons could be very important too. Unfortunately,
there is no standard way to perform such comparisons. Thus,
DAVID only provides annotation results for each of the gene
lists, but does not currently offer any comparison functionality. Statistical tools, such as MS Excel, R, could be very useful
for analysts to conduct these comparisons on the DAVID
results. The comparison may be much customized based on
the analysts statistical knowledge, specific research goals, and
data situation. Several ways to implement the comparisons are
recommended, but are not limited to (1) Directly comparing
the overlapping genes between IL27_Up_List and IFN_Up_
List (Fig. 2). A large number of genes (i.e., 185) are common
between the two independent lists. (2) Particularly, compare
the overlapped genes of the very important and relevant terms
(e.g., interferon induction, antiviral defense, immune response)
that are obtained from step 6 of Chart Reports for IL27_Up_
List and IFN_Up_List, respectively. Interestingly, interferon
inducible genes show over an 80% overlap between the two
lists. (3) Specifically compare the key pathways (e.g., IFN
alpha signaling pathway). The genes of STAT1, STAT2, and
p48 are commonly regulated in both the IL-27 and IFN lists
(Fig. 8). (4) Globally compare the enrichment p values, percentages of gene hits, and/or enrichment fold changes of the
top significantly enriched terms obtained from the DAVID
Chart Report for each of the lists. Analysts should keep in
52
T. Imamichi et al.
Fig. 9. The correlations of enrichment statistical values obtained from DAVID Chart Reports The correlation plots to measure
the annotation agreement between IFN_Up_List and IL27_Up_List are drawn using MS Excel on the data obtained from
Table 1. Regardless two exclusive terms (in yellow circle; more discussion in Table 1), the gene hit percentages and enrichment p values of top enriched terms between the two lists show very strong correlation in overall. It indicates that IFN-
and IL-27 may share, in general, many common mechanisms in MDM treatment experiments.
mind that the statistical values listed in the Chart Reports are
influenced by biology within the gene lists, as well as by other
factors, such as the size of the gene lists (12). Thus, caution
should be used when comparing those statistical values across
gene lists. For IFN_Up_List and IL27_Up_List, a positive
indication for implementing the comparison is that the sizes
of the two lists are fairly consistent. The top enriched terms
obtained from the DAVID Chart Reports largely agree with
each other, which suggests that the two gene lists not only
share common mechanisms for particular terms as discussed
in comparison NO. 2 and 3, but also share many other mechanisms in a global scope (Table 2 and Fig. 9). In contrast, a
similar strong annotation agreement is not observed when
comparing annotation results of two unregulated gene lists
obtained from IL-27 and IFN- treated CD4 T cells (data
not shown). All together, a conclusion could be made that
IL-27 may function through similar mechanisms to that of
IFN- in MDM, but not in CD4 T cells.
References
1. Pflanz, S., Timans, J.C., Cheung, J., Rosales,
R., Kanzler, H., Gilbert, J., Hibbert, L.,
Churakova, T., Travis, M., Vaisberg, E.,
Blumenschein, W.M , Mattson, J.D., Wagner,
J. L., To, W., Zurawski, S., McClanahan, T.K.,
Gorman, D.M., Bazan J.F., de Waal Malefyt,
R., Rennick, D., Kastelein, R.A. (2002) IL-27,
a heterodimeric cytokine composed of EBI3

and p28 protein, induces proliferation of naive
CD4 (+) T cells. Immunity 16: 779790.
2. Hunter, C.A. (2005) New IL-12-family members:
IL-23 and IL-27, cytokines with divergent functions. Nat. Rev. Immunol. 5: 521531.
3. Foli, A., Saville, M.W., Baseler, M.W., Yarchoan,
R. (1995) Effects of the Th1 and Th2 stimulatory cytokines interleukin-12 and interleukin-4
4.
5.
6.
7.
8.
on human immunodeficiency virus replication.

Blood 85: 21142123.
Fakruddin, J. M., Lempicki, R.A., Gorelick,
R.J., Yang, J., Adelsberger, J.W., GarciaPineres, A.J., Pinto, L.A., Lane, H.C.,
Imamichi, T. (2007) Noninfectious papilloma virus-like particles inhibit HIV-1 replication: implications for immune control of
HIV-1 infection by IL-27. Blood 109:
18411849.
Imamichi, T., Yang, J., Huang, D.W., Brann,
T.W., Fullmer, B.A., Adelsberger, J.W.,
Lempicki, R.A., Baseler, M.W., Lane, H.C.
(2008) IL-27, a novel anti-HIV cytokine,
activates multiple interferon-inducible genes
in macrophages. AIDS 22: 3945.
Langer, J.A., Cutrone, E.C., Kotenko, S.
(2004) The class II cytokine receptor (CRF2)
family: overview and patterns of receptor-ligand
interactions. Cytokine Growth Factor Rev. 15:
3348.
Pestka, S., Langer, J.A., Zoon, K.C., Samuel,
C.E. (1987) Interferons and their actions.
Annu. Rev. Biochem. 56: 727777.
Galligan, C.L., Murooka, T.T., Rahbar, R.,
Baig, E., Majchrzak-Kita, B., Fish, E.N. (2006)
Interferons and viruses: signaling for supremacy. Immunol. Res. 35: 2740.
53
9. Samuel, C.E. (2001) Antiviral actions of interferons. Clin. Microbiol. Rev. 14: 778809.
10. Dennis, G. Jr., Sherman, B.T., Hosack, D.A.,
Yang, J., Gao, W., Lane, H.C., Lempicki, R.A.
(2003) DAVID: Database for Annotation,
Visualization, and Integrated Discovery.
Genome Biol. 4: R60.
11. Huang, D.W., Sherman, B.T., Tan, Q., Kir, J.,
Liu, D., Bryant, D., Guo, Y., Stephens, R.,
Baseler, M.W., Lane, H.C., Lempicki, R.A.
(2007) DAVID Bioinformatics Resources:
Expanded annotation database and novel
algorithms to better extract biology from large
gene lists. Nucleic Acids Res. 35: W169W175.
12. Huang, D.W., Sherman, BT., Lempicki, R.A.
(2009) Bioinformatics enrichment tools: paths
toward the comprehensive functional analysis of
large gene lists. Nucleic Acids Res. 37: 113.
13. Lim, W.K., Wang, K., Lefebvre, C., Califano,
A. (2007) Comparative analysis of microarray
normalization procedures: effects on reverse
engineering gene networks. Bioinformatics 23:
i282i288.
14. Harr, B., Schltterer, C. (2006) Comparison of
algorithms for the analysis of Affymetrix
microarray data as evaluated by co-expression
of genes in known operons. Nucleic Acids Res.
34: e8.
Chapter 4
Quantitative Analysis of miRNA Expression in Epithelial Cells
and Tissues
Markus Bitzer, Wenjun Ju, Xiaohong Jing, and Jiri Zavadil
Abstract
Reliable detection of the microRNA (miRNA) precursor and mature form expression levels is a fundamental starting block for more focused studies of the biogenesis and functional roles of these important posttranscriptional modulators of gene expression. Building on our expertise with miRNA expression programs
downstream of TGF-b/Smad signaling in homeostasis as well as in pathological conditions associated with
epithelial tissues, we present a series of detailed and broadly applicable protocols for expression profiling of
the mature miRNA forms using quantitative real-time PCR TaqMan, both single assays or low-density
arrays. We next highlight key steps necessary for the detection of primary precursors of miRNAs
(pri-miRNAs) to address the initial steps of miRNA biogenesis, and we finally review some most widely
used computational algorithms for miRNA target prediction used to complement experimental identification of the target mRNAs and proteins.
Key words: microRNA, miRNA, miRNA expression, Pri-miRNA expression, Quantitative real-time
PCR, TaqMan Array MicroRNA Card, TaqMan MicroRNA Assay, Primer3, SYBR Green PCR,
miRNA target prediction, TGF-b, Epithelial cell, Epithelial tissue
Abbreviations
FFPE
LCM
miRNA
pri-miRNA
qrt-PCR
RNAi
RT
TLDA
Formalin-fixed, paraffin-embedded
Laser capture microdissection
microRNA
Primary precursor of miRNA
Quantitative real-time polymerase chain reaction
RNA interference
Reverse transcription/transcriptase
TaqMan low-density array(s)
55
56
M. Bitzer et al.
1. Introduction
Noncoding regulatory miRNAs are important pleiotropic posttranscriptional modulators of gene expression and function that act
through the RNA interference (RNAi) machinery to inhibit protein
translation initiation and to negatively regulate the mRNA stability
(1). By computational predictions and functional experiments,
miRNAs have been shown to directly regulate at least one half of all
human genes (2, 3). As important modulators of gene expression
programs (4), they play important roles in a variety of fundamental
cellular processes such as the control of cell proliferation, differentiation, and cell death, in the contexts of development and homeostasis
(1, 5). Deregulated miRNAs are also directly involved in the development and progression of a variety of human diseases including
cancers. Recent miRNA-related research thus focuses not only on
understanding of their physiological and pathological roles, but also
on their potential to be used as therapeutical targets or agents.
The pleiotropic cytokine TGF-b is a major player in development, in adult tissue homeostasis, and has been implicated in numerous human diseases. In epithelial cells, TGF-b elicits downstream
transcriptional responses including the expression of miRNAs that in
turn direct target genes with roles in epithelial homeostasis and can
also become deregulated in disease states such as fibrogenesis and
carcinogenesis (6, 7). It is thus important to study the miRNA
expression patterns to understand their contribution to the control
of physiological states or to aberrant disease programs.
Using our expertise derived from long-term studies of the
mammalian epithelial cell and tissue systems with active TGF-b
signaling component, we present a methodological overview of
expression profiling of mature miRNA forms of 2022 nt using
quantitative single TaqMan assays or TaqMan low-density arrays
(TLDA), and we also focus on very important detection of primary
precursors of miRNAs (pri-miRNAs) by quantitative real-time
PCR. The latter methodological approach provides insight into the
first steps of transcriptional control of miRNA expression and biogenesis. Finally, we also review selected computational tools for
miRNA target mRNA prediction. Importantly, all techniques overviewed here can be also applied broadly to the examination of
miRNA roles in any mammalian (mainly human and rodent) tissue
of interest.
2. Materials
2.1. Equipment
1. FlashPAGE Fractionator.
2. Centrifuge with a swing out rotor, Sorvall or Heraeus.
Quantitative Analysis of miRNA Expression in Epithelial Cells and Tissues
57
3. Four centrifuge buckets and ABI card holders (specific to the

Sorvall or Heraeus centrifuge).
4. PCR cycler accommodating 96-well plates, 0.2 mL tubes or
strips.
5. Applied Biosystems 7900HT Fast Real-Time PCR System with
the 96-well, 384-well, and Micro Fluidic Card Upgrade.
6. TaqMan Array Micro Fluidic Card Sealer.
2.2. Reagents
1. Applied Biosystems:
TaqMan MicroRNA Reverse Transcription Kit, 200 reactions.
Multiscribe RT.
dNTPs with dTTP (100 mM) MultiScribe Reverse Transcriptase
(50 U/mL).
10 RT Buffer.
MgCl2 (25 mM).
RNase Inhibitor (20 U/mL).
Megaplex RT Primers (10 ).
Megaplex PreAmp Primers.
TaqMan PreAmp Master Mix.
TaqMan Universal PCR Master Mix No AmpErase UNG.
TaqMan (TLDA) microRNA arrays, human A v2 B v3, or
rodent A and B v2.
2. Nuclease-free water.
3. Oligonucleotide primers (see Subheading 4.1).
4. 96- or 384-Well plates and optical cover (Applied Biosystems).
5. SuperScript First-Strand Synthesis System for RT-PCR
(Invitrogen).
6. Power SYBR Green PCR master mix (Applied Biosystems).
2.3. Software
7900 SDS Software v2.1 or later. Note: SDS Software v2.1 through
v2.2.2 includes the DDCT Study program. SDS Software v2.3
includes the RQ Manager program. Both programs are provided for
relative quantification analysis. Version 2.4 was recently released but
has not been tested by our groups at the preparation of this chapter.
3. Methods
3.1. Isolation of Total
RNA for miRNA
Analysis
The individual and Megaplex RT TaqMan miRNA assays are

designed to work with total RNA isolated by any method which
retains the small RNA contents (<100 nt). There are multiple
commercial extraction kits available for this purpose, of which we
58
M. Bitzer et al.
would like to point out two with which we have a direct experience. First, we recommend the Qiagen miRNeasy series (Mini Kit,
96 Kit for the use of 96-well format), or FFPE Kit for isolation of
total RNA from formalin-fixed, paraffin-embedded tissue sections.
Second, good results are obtained using the miRVana miRNA isolation kit from Ambion (Life Technologies, Inc.) which utilizes
glass fiber filter-based method to isolate total RNA ranging from
10 nt to kilobases in size. These two examples generally yield good
quality total RNA with preserved miRNA contents (typically 0.01%
of the total RNA amount).
3.2. Enrichment of
Small RNA Population
from Total RNA
Isolates
Although routinely not necessary for miRNA profiling described

herein (even when extracting from low-yield sources such as FFPE
histological sections), a fractionation option that enriches for the
small RNA can be added as part of either the abovementioned
Qiagen miRNeasy or Ambion mirVana protocols. A protocol has
also been developed by Ambion using the flashPAGE Fractionator
that loads cell lysates or Trizol-based RNA lysates directly onto the
PAGE column to isolate 1040 nt long nucleic acids, which are in
turn recovered by overnight precipitation at 20C with 1:10 volume of 3 M sodium acetate and 4 volumes of ethanol in the presence of precipitation carrier such as linear acrylamide or molecular
biology grade glycogen. For further information on the enrichment techniques, we refer the reader to the corresponding online
support information of Qiagen and Applied Biosystems (Life
Technologies, Inc.) companies.
3.3. Quantitative
miRNA Expression
Profiling Using
TaqMan Megaplex
Pools
The Applied Biosystems TaqMan Megaplex RT Primers are predefined pools of up to 381 reverse transcription primers for the
one-tube, megaplex reverse transcription of mature miRNA forms.
Two pools of Megaplex RT Primers (Human or Rodent Pools
A and B) are available and are complemented by the respective
microfluidic cards with presynthesized TaqMan probes (see below).
Increased sensitivity of the TaqMan miRNA profiling using limited
RNA content samples (as low as 1 ng total RNA) can be ensured
by a preamplification step using Megaplex PreAmp Primers.
The TaqMan Megaplex pools and microfluidic cards with presynthesized TaqMan primer probes enable medium-to-highthroughput quantitative analysis of hundreds of human and rodent
miRNAs. Currently, the comprehensive coverage of Sanger miRBase v14 database is offered by a two-card set of TaqMan Array
MicroRNA Cards (Cards A and B) for a total of 754 unique assays
specific to human miRNAs, or reflecting the miRBase v10 coverage of 518 mouse and 313 rat miRNA species. In addition, each
card contains a variety of control TaqMan assays to be used as
internal reference controls, including endogenous and negative
control assays.
In general, for either human or rodent samples, the Card A
focuses on more highly characterized miRNAs, while Card B
59
contains many of the more recently discovered miRNAs along with

the existing or predicted miR* (or 3p) sequences. The content
has been upgraded recently to the most recent human Card A v2
and Card B v3, and the rodent Cards A and B v2 (with rodent v3
cards available in 2011). Cards A and B can be purchased and are
also run separately. Overall, the qualitative nature of the TaqMan
MicroRNA Assays, high sensitivity, high specificity, and broad
dynamic range of the methodology, makes this application a better
quality alternative to microarray based profiling techniques.
There are principally two workflows to prepare DNA from
miRNA for TaqMan profiling using the TLDA microfluidic cards.
One is performed without preamplification of the RT product and
is used for robust samples (3501,000 ng total RNA in up to 3 mL,
with a corresponding ~0.01% miRNA contents), and the other is
conducted with preamplification step for samples with low contents of miRNA (FFPE RNA isolates, embryonic explants, LCMdissected samples; in general for yields between 1 and 350 ng total
RNA in up to 3 mL, with ~0.01% contents of miRNA). These distinct workflows are described separately in Subheadings 3.3.2 and
3.3.3. However, the Megaplex RT protocol is a shared starting
point for either workflow:
1. Prepare the RT reaction mix in a 1.5-mL microcentrifuge tube:
3.3.1. Megaplex RT 1
RT reaction mix components
Volume for one

sample (mL)
Megaplex RT Primers (10 )
0.80
dNTPs with dTTP (100 mM)
0.20
MultiScribe Reverse Transcriptase (50 U/mL)
1.50
10 RT Buffer
0.80
MgCl2 (25 mM)
0.90
RNase Inhibitor (20 U/mL)
0.10
Nuclease-free water
0.20
Total
4.50
These volumes are determined for one sample/reaction. Multiply

by N 1.125* for N sample/reactions (* = 12.5% extra volume to
compensate for losses caused by pipetting)
2. Invert the tube(s) 68 times to mix, then centrifuge briefly.
3. In a 96-well plate, 8-tube strips or individual 0.2 mL PCR
tubes, pipette 4.5 mL of each RT reaction mix into each well or
each tube.
Parts of Subheadings 3.3.13.3.3 are adapted from the Applied Biosystems

Megaplex Pools For microRNA Expression Analysis Protocol.
60
M. Bitzer et al.
4. Add 3 mL (0.11,000 ng) total RNA into each well or each

tube containing RT reaction mix.
5. Seal the plate or cap the tubes, invert 68 times to mix, and
spin briefly.
6. Incubate the plate or tubes on ice for 5 min.
7. Set up the PCR run method:
Ramp speed set to standard or maximum.
Reaction volume (mL) = 7.5.
Thermal-cycling conditions:
40 Cycles of:
16C
2 min
42C
1 min
50C
1s
Followed by:
Hold 85C
5 min
Hold 4C
8. Optional stopping point can be inserted here: The cDNA can

now be stored at 20C for at least 23 weeks.
3.3.2. Proceed to PCR
Without Preamplification
(for the Initial 3501,000 ng
RNA Input)
Prepare the PCR mix in a 1.5-mL microcentrifuge tube:

Component
Volume (mL) for one arraya
TaqMan Universal PCR Master Mix

No AmpErase UNG (2 )
Megaplex RT product
450
6
Nuclease-free water
444
Total
900
Includes 12.5% excess for volume loss caused by pipetting errors

Invert the tubes to mix, and then centrifuge the tubes briefly.
Continue to Subheading 3.3.4. Filling the TLDA.
3.3.3. Post-RT
Preamplification Step
for Low-Yield Samples
(0.1350 ng RNA Input)
1. Prepare the PreAmp reaction mix in a 1.5 mL microcentrifuge

tube:
PreAmp reaction mix
Volume for one sample (mL)
TaqMan PreAmp Master Mix (2 )
12.5
Megaplex PreAmp Primers (10 )
2.5
Nuclease-free water
7.5
Total
22.5
61
These volumes are determined for one sample/reaction. Multiply by N

1.125* for N samples/reactions (* = 12.5% extra volume to compensate for losses by pipetting)
2. Invert the tube 68 times to mix and then centrifuge the tubes
briefly.
3. In a 96-well plate or 8-tube strips, pipette 2.5 mL of each RT
product (from step 8 of Subheading 3.3.1) into each corresponding well or tube.
4. Pipette 22.5 mL of PreAmp reaction mix into each well of the
96-well plate or 8-tube strips, or individual 0.2 mL PCR tubes
containing the RT product.
5. Seal the plate (or cap the tubes), invert 68 times to mix, and
spin briefly.
6. Incubate the plate or tubes on ice for 5 min.
7. Set up the PCR run method:
Ramp speed set to standard or max.
Reaction volume (mL): 25.
Thermal-cycling conditions:
Hold 95C
10 min
Hold 55C
2 min
Hold 72C
2 min
12 Cycles of:
95C
15 s
60C
4 min
Followed by:
Hold 99.9C
10 min (required for inactivation of the enzyme)
Hold 4C
8. Remove the 96-well plate or 8-tube strips or individual tubes

from the PCR cycler.
9. Briefly centrifuge the tubes (or plate).
10. Add 75 mL of 0.1 TE, pH 8.0 (or nuclease-free water) to
each well or tube.
11. Seal the plate or cap tubes, then invert six times to mix, spin
briefly.
12. Optional stopping point can be inserted here: In our hands, the
diluted preamplified product can be stored at 20C for at
least 23 weeks.
62
M. Bitzer et al.
Prepare the PCR mix in a 1.5 mL microcentrifuge tube:

Component
Volume for one arraya
TaqMan Universal PCR Master Mix

No AmpErase UNG (2 )
450
Diluted PreAmp product
Nuclease-free water
441
Total
900
Includes 12.5% excess for volume loss from pipetting

Invert the tubes to mix and centrifuge them briefly.
Continue below for instructions on how to load and run the array
using the 384-well TLDA default thermal-cycling conditions.
3.3.4. Filling the TaqMan
Arrays2
1. Allow the TLDA card to equilibrate to room temperature for

at least 10 min, and then remove it from its packaging.
2. Place the TLDA on a clean laboratory bench surface (not ice)
with the foil bottom facing down.
3. Load a 100 mL aliquot of the 900 mL sample-specific PCR mix
into a 100 mL micropipette.
4. Hold the micropipette at an angle and insert the tip in the first
fill port. The fill port on the left arm of each fill reservoir is the
larger of the two openings (see Fig. 1).
5. Pipet carefully the sample-specific PCR mix so that it flows in
and around the fill reservoir in the direction of the smaller
(vent) port.
6. Fill the remaining ports with remaining 100 mL aliquots of the
same PCR mix.
Fig. 1. Filling the TLDA with 100 mL of PCR Mix per fill port.
2
Parts of Subheadings 3.3.4 and 3.3.7 are adapted from the Applied Biosystems
TaqMan Array User Bulletin.
3.3.5. Centrifuging
the TaqMan Arrays
63
After all eight fill reservoirs are filled with the PCR mix, centrifuge
the TaqMan Array card to deliver the mix to the 384 reaction wells
with the presynthesized TaqMan probes. Sorvall or Heraeus centrifuge with the Sorvall/Heraeus Applied Biosystems Custom
Buckets and TLDA card holders must be used exclusively.
1. Position the bucket on a bench, so the label faces you.
2. Insert carefully the TaqMan Array cards into the card holder,
ensuring that the fill reservoirs project above the card holder. The
TLDA card side with reaction wells faces the same direction as
the label on the card holder and on the bucket.
3. Always use blank balance cards provided with the installation
kit to fill any unused positions in the card holder. The centrifuge rotor holds four Sorvall/Heraeus buckets. Each bucket
holds up to three TaqMan Array cards (loaded and/or blank
balance cards) in the card holder. The card holder supports the
TaqMan Array card during centrifugation.
Note: Using blank balance cards to fill any open positions prevents potential damage to the card holder. If the card holder is
not completely filled, the TaqMan Array card may become displaced during centrifugation, resulting in uneven filling and
loss of precious data.
4. Place a filled card holder in the bucket so that the This Side
Out label faces the front of the bucket.
5. Centrifuge under these spinning conditions:
Acceleration ramp rate 9.
Deceleration ramp rate 9.
Rotational speed 1,200 rpm (331 g).
Centrifugation time 2 1 min.
(Check the progress between these two 1 min spins by inspecting the cards wells for proper filling).
3.3.6. Sealing of the

Centrifuged TaqMan Arrays
The TaqMan Array Micro Fluidic Card Sealer (see Fig. 2) physically separates the wells of a TaqMan Array card filled with the
PCR mix. The sealer creates a physical barrier along the main fluid
distribution channels of the TaqMan Array card. The manufacturer
recommends using a slow, even, and deliberate motion when operating the sealer, which is a key to the successful application of the
TaqMan Array cards.
1. Place the sealer on a firm bench, orient it such that the front
end (the starting position shown to the left on Fig. 2) is
closest to you and the opposite end is farthest from you. The
correct position is also indicated by the arrows on the sealer
pointing away from you. Move the sealers carriage in its starting
position, i.e., closer to you.
64
M. Bitzer et al.
Fig. 2. The TaqMan Array Micro Fluidic Card Sealer with the filled TLDA loaded for
sealing.
2. Insert a TaqMan Array card into the sealer by orienting it in the

right direction to fit the sealers insert compartment. The cards
fill reservoir should be positioned closest to the arrows on the
sealers base. Align the cards pin grooves, the well side down
and foil side up, to the pins on the sealer. With care, place the
card on top of the insert compartment and make sure that the
cards front end is held securely in place by the spring clips.
Push the card until it slips completely and securely into the
insert compartment. When properly inserted, the cards foil
surface is level with the base of the sealer. The four spring clips
hold the card in correct position.
3. Firmly hold and slide the carriage over the base of the sealer in
the direction of the arrows (pushing away from you). Apply a
slow, uninterrupted motion to move the carriage across the
entire length of the TaqMan Array card, until the carriage
reaches the stops at the far end which prevent it from coming
off. Refrain from moving the carriage too quickly and with
excessive force. Do not return the carriage to the starting position
before completely removing the sealed TaqMan Array card.
4. To remove the sealed card for further processing, lift it off
the sealers insert compartment by using the thumb slots that
provide easy access to the sides of the card.
5. Inspect the TaqMan Array card to ensure that the sealing was
successful. The indentations from the sealers precision stylus
assembly should exactly align with the cards main distribution
channels.
6. Cut away and discard the fill strip of the TaqMan Array card
using scissors guided along the edge of the cards plastic carrier
(see Fig. 3). The card is now prepared to be processed on the
7900HT SDS equipment.
65
Fig. 3. Cutting-off the fill strip by scissors, using the edge of the plastic cards carrier.
3.3.7. Running the Cards

Using the SDS2.3 Software
on the ABI 7900HT SDS
Cycler
To collect and analyze the miRNA TLDA data using the ABI
7900HT SDS equipment, the user will need to use the Applied
Biosystems 7900 software containing two components the
SDS2.3 software (performs most real-time analysis related tasks
such as collecting data, setting up templates, and allowing data
analysis using standard curves, but will not allow analysis of DDCt
RQ runs described below) and the RQ Manager 1.2 software.
We recommend running the plates under these settings: Assay
type: DDCt (RQ) and Container: 384-well TLDA (these options
have to be chosen at the start of each run). The run template is
provided on a CD with each card as a text tab-delimited sample
annotation file (SDS Setup file) containing the SDS extension
(some examples for latest human cards are SDS_miRNA_Human_A.
txt or HumanMiRNAPoolB_V3_SDS.txt). This file is simply
imported to provide all well-specific annotations as part of the
template. The Sample name corresponding to one single sample
per card for miRNA analysis needs to be entered separately and
manually for the entire plate.
The RQ Manager component of the 7900 software allows to
analyze data from up to 10 DDCt RQ runs simultaneously, as relative quantification templates, using the 2DDCt formula to calculate
final relative quantities of a particular miRNA across studied samples. In order to achieve the best software performance and analysis results, we recommend to installing the ABI-provided software
SDS_v2.3_rev_C_patch on top of the basic SDS/RQ installation.
This patch is distributed through the Support pages of Applied
Biosystems.
While the RQ Manager is capable of overlaying multiple plates
and generates consistent assay-specific thresholds, and generally
well-established RQ values normalized to select endogenous
controls, some functionality is missing such as averaging across
66
M. Bitzer et al.
replicate experiments for the same condition or defining multiple

samples as experiment baseline. In such case, the investigator might
want to export the Ct values and process them outside the RQ
Manager software in a tabular data processing software such as
Microsoft Excel, using the 2DDCt formula (8), or import the processed relative quantity data into a sophisticated analysis platform
such as Agilents GeneSpring GX11.5 software. The details of these
customized normalizations and analyses are, however, beyond the
scope of this chapter.
4. Immature
miRNA Precursor:
Detection by qPCR
4.1. Primer Design
miRNAs are processed from primary precursor miRNAs (primiRNA) and it has been shown that the level of active, mature
miRNAs is regulated through transcriptional regulation of the pricursor and/or through modulation of the maturation process. To
distinguish these very different regulatory mechanisms, it is necessary not only to sensitively quantify levels of the mature miRNA
but also those of the associated pri-miRNAs.
Pri-miRNAs are transcribed by RNA polymerase II from
miRNA genes located within introns of coding and noncoding
genes, exons of noncoding genes, or in intergenic regions. Pricursors have all the characteristics of mRNAs including 5 cap and
polyA tail and both criteria are used to identify the genomic location of pri-miRNAs. Pri-miRNAs are larger than 300 nt and can
vary in size; in particular, miRNA genes located in intergenic
regions can be several kilobases long. Adjacent promoter regions
suggest transcriptional regulation of primary transcript that follows
conventional rules for mRNAs. Therefore, pri-cursor expression
can be determined using standard techniques used for mRNA. We
are using the Applied Biosystems 7900HT Real-Time PCR System
with 96- and 384-well block and corresponding plates; the provided protocol is designed for use with this system. If a different
system is used, the protocol may need to be modified.
When designing the primers for pri-miRNAs, it is important that at
least one primer binds outside of the precursor sequence. For those
pri-miRNAs whose sequence are currently available, primers can
be designed using Primer3 program (as described in the next paragraph). For those unknown pri-miRNA, it is more challenging to
experimentally confirm and characterize. An acceptable approach is
to derive the sequence of its relevant pre-miRNA through miRBase (9) (http://www.mirbase.org) and use it to blast the genome.
The predicted but incomplete pri-cursor sequence for this premiRNA is obtained by adding 100200 nt to both 5 and 3 end of
67
the pre-miRNA sequence. This sequence can be used to design

conventional qrt-PCR primers.
This approach does not detect isoforms of mature miRNAs
that differ sometimes in only a single nucleotide. In addition, one
primary transcript can contain several mature miRNAs.
4.2. Reverse
Transcription
Go to: http://www.mirbase.org and search for the miRNA of

interest.
Select species and type gene symbol of miRNA of interest.
Copy sequence of stem loop (=precursor) into blast function

of genome browser (i.e., http://genome.ucsc.edu/cgi-bin/
hgBlat or http://blast.ncbi.nlm.nih.gov/Blast.cgi).
Blast your stem loop sequence and obtain genomic sequence

approximately 200 bp upstream of 5 end and downstream of
3 end of precursor. If a transcript that includes the precursor
sequence is available, it can be used to design of primers.
Design primers for quantitative real-time PCR. Primer pairs

can be picked and optimized for conditions using publicly
available software for primer design (i.e., Primer3 http://
frodo.wi.mit.edu/primer3/). The PCR product length was
designed to be 80120 bp.
The SuperScript First-Strand Synthesis System for RT-PCR from

Invitrogen is used for reverse transcription and the provided protocol is slightly modified based on SuperScript II Reverse
Transcriptase manufactorys instructions.
1. Prepare the following RNA/primer mixture in each tube:
Total RNA
1 ng5 mg (we routinely use 1 mg,

it is recommended to use the
same amount of total RNA for
all experimental samples
whenever possible)
Oligo(dT)1218 (500 mg/mL):
1 mL
10 mM dNTP mix
1 mL
RNase/DNase-free H2O
10 mL
2. Incubate the samples at 65C for 5 min and then on ice for at
least 1 min.
3. Prepare reaction master mixture. For each reaction:
5 First-Strand Buffer
4 mL
0.1 M DTT
2 mL
RNAaseOUT
1 mL
68
M. Bitzer et al.
4. Add the reaction mixture to the RNA/primer mixture, mix

briefly, incubate at 42C for 2 min.
5. Add 1 mL (200 U) of SuperScript II RT and mix by pipetting
gently up and down.
6. Incubate the tubes at 42C for 50 min, heat inactivate at 70C
for 15 min, and then chill on ice.
7. Collect sample by brief centrifugation, add 1 mL RNase H, and
incubate at 37C for 20 min. This step is optional and may be
considered if larger amounts of RNA (>1 mg) are used.
8. Store the first-strand cDNA at 20C until use for real-time
PCR.
4.3. Real-Time PCR
1. Normalize the primer concentrations and mix gene-specific forward and reverse primer pair. Each primer (forward or reverse)
concentration in the mixture is 2.5 pmol/mL (2.5 mM).
2. The total volume of the real-time PCR mixture can be modified depending on the plate (96- or 384-well plate). Minimum
volume is preferred due to economic considerations. In our
hands, minimal volume of 20 mL for 96-well plate and 10 mL
for 384-well plate yield reliable results. Larger total volumes can
be considered to increase robustness (evaporation or pipetting
contributes relative more imprecision in small volumes than in
lager volumes).
For 10 mL:
5 mL SYBR Green Mix (2).
2 mL diluted cDNA (cDNA can be diluted 20 to 40 using
DNase- and RNase-free H2O).
2.5 mL primer pair mix (2.5 pmol/mL each primer).
0.5 mL H2O.
(Double each volume for 20 mL reaction volume)
3. Standard PCR program includes:
Hold 50C
2 min
Hold 95C
10 min
40 Cycles of:
95C
15 s
60C
60 s
Followed by:
Hold 72C
10 min
4. After PCR is finished, remove the tubes from the machine. The
PCR specificity is examined by either 3% agarose gel using
5 mL from each reaction or by dissociation curve analysis.
69
5. Analyze the real-time PCR result with the SDS 7900 software
(see Subheading 3.3.7.) or in Excel spreadsheet using the using
the 2DDCt formula (8). Always inspect the amplification traces
to see if there is any bimodal dissociation curve or abnormal
amplification plot.
5. Algorithms for
Prediction of
miRNA Target
Genes
As the function of miRNAs is determined by the regulatory effects

on the function of target genes, many different computational algorithms have been generated that predict target genes based on different parameters, and essentially all include sequence homology to
the seed sequence (usually base 28 at the 5 end) as one of their
parameters. Only a very limited number experiments have been
published so far that attempt to determine the accuracy of these
predictions. These are largely based on unbiased transcriptomic and
proteomic methods of changes in protein expression after manipulation of individual miRNAs in cell culture systems (10, 11). More
recently, experimental approaches using sequencing of miRNAs
and mRNA fragments co-immunoprecipitating with Argonaut proteins of the RISC complex have been used. But none of these methods identifies direct miRNA targets independent of certain sequence
homologies. Of those algorithms with different provenance yet
comparable performance, a relatively stringent and regularly updated
algorithm is provided by the TargetScan tool (2) (http://www.targetscan.org) that also ranks the targets for each miRNA by a context score, considering the sequence contents in the vicinity of the
target site. Similarly performing algorithm is PicTar (http://pictar.
mdc-berlin.de/) (12). Experimental data have validated the ranking order of the scores as a measure of prediction reliability (11).
A less stringent prediction program is miRanda (13, 14) that provides a much larger but likely less reliable number of predicted targets for each miRNA than TargetScan, similarly to MicroCosm
Targets (formerly miRBase Targets) developed by the Enright
group as part of the miRBase database (9, 15). RNA22 is an algorithm that allows the user to modify the settings used to search for
target genes (16). It has to be emphasized that the overlap is limited
between predictions from different algorithms, with an exception
of PicTar and TargetScan (10, 11). Therefore, the obtained results
have to be used with caution. Currently, probably no single computational system is available to predict target genes robustly and
without errors and thereby infer the functions of upstream miRNA
regulators. The concordance and the validation of performance
between TargetScan and PicTar make them a widely used and recommended set of tools for target prediction, while experimental
functional validations of the target predictions are usually required.
70
M. Bitzer et al.
References
1. Bartel, D.P. (2004) MicroRNAs: genomics,
biogenesis, mechanism, and function. Cell
116: 281297.
2. Friedman, R.C., Farh, K.K., Burge, C.B.,
Bartel, D.P. (2009) Most mammalian mRNAs
are conserved targets of microRNAs. Genome
Res. 19: 92105.
3. Lim, L.P., Lau, N.C., Garrett-Engele, P.,
Grimson, A., Schelter, J.M., Castle, J., Bartel,
D.P., Linsley, P.S., Johnson, J.M. (2005)
Microarray analysis shows that some microRNAs downregulate large numbers of target
mRNAs. Nature 433: 769773.
4. Bartel, D.P. (2009) MicroRNAs: target recognition and regulatory functions. Cell 136:
215233.
5. Croce, C.M. (2009) Causes and consequences
of microRNA dysregulation in cancer. Nat.
Rev. Genet. 10: 704714.
6. Massagu, J. (2008) TGFbeta in Cancer. Cell
134: 215230.
7. Zavadil, J., Bttinger, E.P. (2005) TGF-beta
and epithelial-to-mesenchymal transitions.
Oncogene 24: 57645774.
8. Livak, K.J., Schmittgen, T.D. (2001) Analysis
of relative gene expression data using real-time
quantitative PCR and the 2(Delta Delta
C(T)) Method. Methods 25: 402408.
9. Griffiths-Jones, S., Grocock, R.J., van Dongen,
S., Bateman, A., Enright, A.J. (2006) miRBase:
microRNA sequences, targets and gene nomenclature. Nucleic Acids Res. 34: D140144.
10. Baek, D., Villn, J., Shin, C., Camargo, F.D.,

Gygi, S.P., Bartel, D.P. (2008) The impact of
microRNAs on protein output. Nature 455:
6471.
11. Selbach, M., Schwanhusser, B., Thierfelder,
N., Fang, Z., Khanin, R., Rajewsky, N.
(2008) Widespread changes in protein synthesis induced by microRNAs. Nature 455:
5863.
12. Krek, A., Grn, D., Poy, M.N., Wolf, R.,
Rosenberg, L., Epstein, E.J., MacMenamin,
P., da Piedade, I., Gunsalus, K.C., Stoffel, M.,
Rajewsky, N. (2005) Combinatorial microRNA
target predictions. Nature Genet. 37:
495500.
13. Betel, D., Wilson, M., Gabow, A., Marks, D.S.,
Sander, C. (2008) The microRNA.org
resource: targets and expression. Nucleic Acids
Res. 36: D149153.
14. John, B., Enright, A.J., Aravin, A., Tuschl, T.,
Sander, C., Marks, D.S. (2004) Human
MicroRNA targets. PLoS Biol. 2: e363.
15. Griffiths-Jones, S., Saini, H.K., van Dongen, S.,
Enright, A.J. (2008) miRBase: tools for
microRNA genomics. Nucleic Acids Res. 36:
D154158.
16. Miranda, K.C., Huynh, T., Tay, Y., Ang, Y.S.,
Tam, W.L., Thomson, A.M., Lim, B.,
Rigoutsos, I. (2006) A pattern-based method
for the identification of MicroRNA binding
sites and their corresponding heteroduplexes.
Cell 126: 12031217.
Chapter 5
Evaluating Posttranscriptional Regulation of Cytokine Genes
Bernd Rattenbacher and Paul R. Bohjanen
Abstract
A wide variety of cytokines are necessary for cellcell communication in multicellular organisms, and
cytokine dysregulation has detrimental effects, leading to disease states. Thus, it is a necessity that the
expression of cytokines is tightly controlled. Regulation of cytokine gene expression takes place at different
levels, including transcriptional and posttranscriptional levels. Ultimately, the steady-state levels of cytokine
transcripts are determined by the equilibrium of transcription and degradation of this mRNA. Degradation
rates of cytokine mRNAs can be measured in cells by blocking transcription with actinomycin D, harvesting RNA after different time points, and evaluating mRNA levels over time by northern blot. Cis-acting
elements that mediate the rapid decay of numerous cytokine transcripts, including AU-rich elements
(AREs), are found in the 3 untranslated region (UTR) of these transcripts. Putative regulatory cis-elements
can be cloned into the 3 UTR of a reporter transcript in order to assess their function in regulating mRNA
decay. Cis-elements, such as AREs, regulate cytokine mRNA decay by binding to trans-acting proteins,
such as tristetraprolin or HuR. These RNA-binding proteins can be visualized using electromobility
shift assays or UV crosslinking assays based on their binding to radioactively labeled RNA sequences.
RNA-binding proteins that regulate cytokine mRNA decay can be purified using an RNA affinity method,
using their target RNA sequence as the bait. In this chapter, we review the methods for measuring cytokine
mRNA decay and methods for characterizing the cis-acting elements and trans-acting factors that regulate
cytokine mRNA decay.
Key words: mRNA decay, Actinomycin D chase, Northern blot, RNAprotein interaction, EMSA,
UV crosslinking, One-step affinity purification, AU-rich element, Tristetraprolin, HuR
1. Introduction
Cytokines regulate a variety of different events in the human body.
They provide signals to cells, telling them when to divide, what
proteins to produce, what other cytokines to secrete, and how they
should differentiate. Because of the importance of cytokines in
maintaining homeostasis and the dangers involved when cytokines
71
72
B. Rattenbacher and P.R. Bohjanen
become dysregulated, it is crucial that the expression of cytokines

is tightly regulated at multiple levels, including transcriptional and
posttranscriptional levels. Steady-state mRNA levels are determined
by the balance between transcription and mRNA degradation.
The biochemical mechanisms that regulate the degradation of
cytokine transcripts is not well understood, although there is evidence that failure to degrade pro-inflammatory cytokine transcripts
such as TNF, INF, IL2, IL6, IL8, or IL10 transcripts leads to
chronic inflammation (15). The degradation of cytokine transcripts
is regulated through cis-elements in their 3 untranslated regions
(3 UTRs) and trans-acting factors that bind to them. One important and well-known destabilizing element is the AU-rich element
(ARE), which is found in the 3 UTR of many unstable cytokine
mRNAs (69). AREs function by binding to trans-acting proteins
that regulate the stability of the transcript. Some ARE-binding proteins, such as tristetraprolin and butyrate response factor 1, are
responsible for rapid degradation of ARE-containing transcripts
(1012), whereas other proteins, such as HuR, have the potential
to stabilize the same message (9, 1318). Whether an AREcontaining cytokine transcript undergoes degradation or stabilization is dependent upon the activation status of the cells. Activation
of a cell, for example, will lead to an inactivation of TTP or BRF1
(3, 19) and the recruitment of HuR to the ARE, which will result
in increased stability of the ARE-containing message (9, 1318). In
this way, a cell can respond rapidly to changes in the outside environment and produce the required cytokine.
This chapter introduces the techniques used to measure degradation of cytokine transcripts and to characterize the RNA-binding
proteins involved in their regulation. We have used the techniques
described below to characterize the role of mRNA decay in regulating the expression of cytokine genes and other early response
genes after activation of primary human T cells (7, 8, 20, 21), and
to identify and characterize the function of AREs and ARE-binding
proteins in the regulation of cytokine mRNA decay (1, 10, 13, 14,
17). Although we have used the techniques to study posttranscriptional cytokine gene regulation in human T cells, the techniques
can be broadly applied to a variety of cell types and experimental
systems.
2. Materials
2.1. Measuring mRNA
Decay Rates
2.1.1. Isolation of Primary
Human T Cells
1. 40 mL Buffy coat white blood cell packs of 109 cells or more

(American Red Cross). Buffy coat cells must be used
immediately.
2. 2 mL RosetteSep human T cell enrichment cocktail (StemCell
Technologies, Inc.).
73
3. Dulbeccos Phosphate-buffered saline (PBS, Gibco) supplemented

with 2% (v/v) FBS. Store at 4C.
4. Ficoll-Paque Plus (GE-Healthcare Amersham).
5. Whole blood erythrocyte lysing kit (R&D Systems).
6. RPMI medium 1640 (Gibco) supplemented with 10% (v/v)
FBS, 1% (v/v) penicillin/streptomycin (Gibco), and 1% (v/v)
L-glutamine (Gibco).
2.1.2. T Cell Stimulation
and Measurement
of mRNA Decay After
Addition of Actinomycin D
1. Tissue culture dish with 20 mm grid 150 25 (Falcon) for

T cell stimulation.
2. Anti-hCD3 Antibody (R&D Systems) and anti-hCD28
Antibody (R&D Systems) resuspended in 500 L sterile water.
Aliquot 50 L and freeze at 20C. Make coating solution
fresh and use immediately; add 5 g anti-hCD3 antibody and
5 g anti-hCD28-antibody into 10 mL PBS per plate to be
coated. Antibody solutions will be stable for several months
at 20C.
3. Actinomycin D-mannitol (Sigma) is resuspended in 1 mL sterile PBS. Actinomycin D should not be used after 1 month in
solution. Store at 20C.
4. 70% (v/v) Ethanol.
5. 2-Mercaptoethanol.
6. Qiashredder (Qiagen).
7. RNeasy mini kit (Qiagen).
2.1.3. b-Globin Reporter

Based Assays to Measure
mRNA Decay Half-Lives
1. HeLa Tet-Off cells (Clontech).

2. OptiMEM (Gibco).
3. Lipofectamine 2000 Reagent (Invitrogen).
4. BBB-plasmid (9).
5. pTracer-EF C vector (Invitrogen).
6. TrypLE Express (Gibco).
7. Minimum essential medium alpha (MEM, Gibco) supplemented with 10% (v/v) Tet system approved FBS (Clontech),
1% (v/v) penicillin/streptomycin (Gibco), and 1% (v/v)
L-glutamine (Gibco).
8. Dissolve doxycycline (Clontech) to a concentration of
300 g/L in DMSO. This will result in a 1,000 stock solution. Store at 20C.
9. 2-Mercaptoethanol.
10. Qiashredder (Qiagen).
11. RNeasy mini kit (Qiagen).
74
2.1.4. Northern Blotting
1. NorthernMax-Gly Sample Loading Dye (Ambion).

2. NorthernMax-Gly 10 Gel Prep/Running Buffer (Ambion).
3. BrightStar-Plus
(Ambion).
Positively
Charged
Nylon
Membrane
4. NorthernMax One-Hour Transfer Buffer (Ambion).

5. ULTRAhyb Ultrasensitive Hybridization Buffer (Ambion).
6. NorthernMax Low Stringency Wash Buffer (Ambion).
7. NorthernMax High Stringency Wash Buffer (Ambion).
8. 3 mm Chr Chromatography paper (Whatman).
2.1.5. Probe Preparation
for Northern Blotting
1. Taq DNA-Polymerase set (Qiagen).

2. The low dATP dNTP Mix contains 10 mM dGTP, 10 mM
dTTP, 10 mM dCTP, and 2 mM dATP final concentrations in
RNase-free water (Roche).
3. Primers for -globin probe: -globin+ (5-GTC TAC CCA
TGG ACC CAG AGG-3), -globin (5-AGG ATC CAC
GTG CAG C-3).
4. Primers for GFP probe: GFP+ (5-CCA TGG CTA GCA AAG
GAG-3), GFP (5-CCA TGT GTA ATC CCA GCA GCA
G-3).
5. 32P-dATP (MP Biochemicals).
6. Microspin G-25 Columns (GE-Healthcare).
2.2. Characterization
of RNAProtein
Interactions
2.2.1. Preparation
of Cytoplasmic Extracts
1. NP40 Lysis buffer: 10 mM HEPES/KOH, pH 7.9, 40 mM

KCl, 3 mM MgCl2, 5% (v/v) glycerol, 0.2% (v/v) NP40. NP40
Lysis buffer can be stored at 4C without DTT and inhibitors.
Before cell lysis add 1 mM DTT, 2 ng/mL leupeptin, 2 ng/mL
pepstatin, 8 ng/mL aprotinin, 0.1 mg/mL phenylmethylsulfonyl fluoride (PMSF).
2. Biorad protein assay reagent (Biorad).
2.2.2. Electrophoretic
Mobility Shift Assay
1. RBB buffer: 25 mM HEPES/KOH, pH 7.9, 40 mM KCl,

3 mM MgCl2, 5% (v/v) glycerol, 1 mM DTT, can be stored at
4C or aliquoted and frozen at 20C.
2. Heparan sulfate (Sigma) is reconstituted in 100 L RNase-free
water to obtain a 50-mg/mL solution and should be aliquoted
and stored at 20C.
3. 5 TB-buffer: 54 g Trisbase and 27.5 g boric acid in 1 L of
water. 1 TB-buffer is used for gel preparation and running
buffer. Store at room temperature.
4. 30% (w/v) acrylamide/bis solution, 19:1 (Bio-Rad).
75
5. 6 DNA loading dye: 30% (v/v) glycerol and 0.25% (w/v)

bromophenol blue in RNase-free water. Store at room
temperature.
2.2.3. End-Labeling
of Synthetic RNA Oligomer
Probes
1. Custom-synthesized RNA (Dharmacon) is dissolved as a

100 M solution and is stored at 20C.
2. T4 Polynucleotide Kinase (Invitrogen) which includes 5 forward reaction buffer.
3. 32P-ATP (MP Biochemicals).
4. Phenol/chloroform/isoamyl alcohol (Invitrogen).
5. Prepare a 3-M sodium acetate solution in RNase-free water.
Store at room temperature.
6. Microspin G-25 Columns (GE-Healthcare).
2.2.4. UV Crosslinking
Assays
1. Heparan sulfate (Sigma) is reconstituted in 100 L RNase-free

water to obtain a 50-mg/mL solution and should be aliquoted
and stored at 20C.
2. RNase T1 (Ambion).
3. RBB buffer: 25 mM HEPES/KOH pH 7.9, 40 mM KCl,
4. 5 Sample buffer: 25 mM TrisHCl pH 6.8, 500 mM dithiothreithol (DTT), 0.1% (w/v) bromophenol blue, 50% (v/v)
glycerol. Aliquot and store at 20C.
2.2.5. Probes for UV

Crosslinking Assays
1. Promega kit (Promega).

2. rNTP-mix: combine 10 mM rATP, 10 mM rCTP, and 10 mM
rGTP final concentration in RNase-free water. Store at
20C.
3. 32P-UTP (MP Biochemicals).
4. RNA stop solution mix: 10 mL Tris/EDTA (10 mM/1 mM),
pH 8.0 with 10 L 0.5 M EDTA, pH 8.0, and 100 L 10%
(w/v) SDS. Store at room temperature.
5. Phenol/chloroform/isoamyl alcohol (Invitrogen).
6. Prepare a 3-M sodium acetate solution in RNase-free water.
Store at room temperature.
2.2.6. Isolation
of RNA-Binding Proteins
by One-Step Affinity
Chromatography
1. Custom 5 biotinylated RNA and unbiotinylated competitor

RNA (synthesized by Dharmacon).
2. Dissolve heparan (Sigma) as 100 mg/mL solution in RNasefree water. Aliquot and store at 20C.
3. M-280 Streptavidin beads (Dynal Biotech).
76
4. RBB buffer: 25 mM HEPES/KOH, pH 7.9, 40 mM KCl,

3. Methods
This methods section is divided into two parts. The first part
describes methods to measure decay rates of cytokine mRNAs, and
the second part details methods to characterize RNA-binding proteins that bind to specific decay elements, such as the ARE.
The methods described below all involve RNA in one way or
another. Since RNA is prone to degradation by RNases, which can
be found in dust and sweat, it is crucial that gloves are worn at all
times and that the workspace is clean. To clean the workbench, we
use RNA-ZAP (Ambion). Another source for RNase contamination is water used to make solutions and buffers; therefore, RNasefree water should be used for preparing every buffer and reaction.
RNase-free water can be prepared relatively inexpensively using
DEPC, which will destroy enzymes (reaction with histidines), and
its preparation is described (see Note 1). DEPC needs to be inactivated by thoroughly autoclaving this solution. Failure to inactivate DEPC will result in nonreproducible results.
3.1. Measuring mRNA
Decay Rates
The half-lives of different transcripts expressed in T cells vary

greatly (8). Also, the activation status of T cells can greatly influence the rate of mRNA decay (8). This section focuses on the isolation of T cells, on how to stimulate them and how to prepare total
RNA in order to measure transcript half-lives of the endogenous
cytokine mRNAs. We describe methods for measuring cytokine
mRNA decay rates by northern blot following transcriptional arrest
with the RNA polymerase II inhibitor, actinomycin D.
After it is ascertained that a certain transcript is regulated at the
level of mRNA decay, the cis-elements that are responsible for this
regulation can be identified using a reporter based assay. The most
widely used reporter in the field of mRNA degradation is the
-globin based reporter plasmid, BBB, which produces the -globin
transcript under control of a tetracycline-regulated promoter (9).
The putative degradation sequence can be cloned into the BBB
plasmid via a unique BglII site in the 3 UTR. This plasmid can
then be transfected into HeLa cells expressing a tet-repressor, and
-globin mRNA decay can be measured by northern blot after
turning off transcription by the addition of doxycycline. The advantage of measuring mRNA decay in this system compared to the
actinomycin D system is that doxycycline has minimal effects on
the cells, whereas actinomycin D is extremely toxic. Another
advantage is, that the function of each putative decay element can
77
be singled out and studied individually. The disadvantages of this

system are that reporter transcripts are studied, rather than the
endogenous native transcripts, and the experiments cannot be performed in primary cells.
3.1.1. Isolation of Primary
Human T Cells
1. Separate 40 mL of buffy coat into two 50 mL tubes equally

(see Note 2).
2. Add 1 mL of RosetteSep Human T Cell Enrichment Cocktail
(StemCell Technologies Inc) to each tube.
3. Mix by gently shaking the tubes.
4. Incubate the reaction for 20 min at room temperature to allow
the antibodies to agglutinate everything except CD4-positive
cells.
5. Bring each sample up to 35 mL with PBS supplemented with
2% (v/v) FBS (PBS/FBS) after the incubation is finished.
6. Prepare two new tubes containing 15 mL Ficoll-Paque
Plus.
7. Layer the blood carefully over the Ficoll. Avoid mixing of the
layers at all times (see Note 3).
8. Spin the samples at 1,700 g for 20 min with the brake turned
off.
9. Remove the top layer (serum) carefully to avoid disturbing the
white, cloudy interphase which contains CD4 positive T cells.
10. Transfer the cells in the interphase to two new 50 mL tubes.
11. Add PBS/FBS solution up to 50 mL and centrifuge at 230 g
for 8 min.
12. Remove the supernatant carefully and as completely as possible
without disturbing the cell pellet.
13. Remaining red blood cells are removed with the whole blood
erythrocyte lysing kit (R&D Systems). Resuspend both pellets
in 5 mL 1 lysis buffer and pool them in a 50 mL tube and
incubate for 10 min at room temperature to allow erythrocyte
lysis.
14. Add 40 mL of 1 wash buffer and spin tube at 230 g for
8 min.
15. Remove the supernatant; the pellet should appear white now.
16. Resuspend cells in 50 mL RPMI medium 1640 supplemented
(Gibco, 10% FBS, 1% penicillin/streptomycin (Gibco), 1%
L-glutamine (Gibco)).
17. Count cells using a hemocytometer (use a dilution of 1:100
cell suspension to PBS).
18. Cells are incubated overnight in an incubator at 37C, 5%
(v/v) CO2.
78
3.1.2. T Cell Stimulation

and Measurement
of mRNA Decay after
Addition of Actinomycin D
1. Coat four 15 cm petri dishes overnight at 4C with coating

solution (10 mL PBS containing 5 g anti-hCD3 antibody
and 5 g anti-hCD28 antibody) and four 15 cm dishes with
PBS only. Note: Use one plate for each time point of the actinomycin D chase experiment (see Note 4).
2. Remove coating solution or PBS.
3. Wash plates with 10 mL PBS. Remove PBS.
4. Equilibrate plates with 5 mL RPMI 1640 medium with
supplements.
5. Remove medium.
6. Add 3 107 cells in 15 mL RPMI-medium with supplements
to each plate.
7. Incubate cells for 3 h at 37C and 5% (v/v) CO2.
8. Add actinomycin D at a final concentration of 5 g/mL to
each plate (see Note 5).
9. Incubate the plates at 37C and 5% (v/v) CO2 for 0, 1, 2, or
3 h.
10. Remove cells by scraping. Note: Do not remove medium
before scraping because the medium may contain floating
cells.
11. Transfer cells and medium to a 50 mL tube.
12. Collect cells by spinning at 340 g for 5 min.
13. Remove medium as completely as possible without disturbing
the pellet.
14. Extract total RNA with the RNeasy Mini kit following manufacturers recommendations (see Note 6).
15. Elute total RNA in 50 L RNase-free water.
16. Estimate RNA concentration using a spectrophotometer. The
RNA concentration should be between 0.5 and 3 g/L (see
Note 7).
17. Analyze the stability of the cytokine mRNA by northern blot;
probes for endogenous RNA can be prepared (see Note 8).
3.1.3. b-Globin Reporter

Based Assays to Measure
mRNA Decay Half-Lives
1. HeLa Tet-Off cells are propagated in HeLa Tet-Off medium

(see Note 9).
2. Seed HeLa Tet-Off cells into 15 cm dishes with HeLa Tet-Off
medium; the cells should be about 90% confluent the next
day.
3. Prepare a Lipofectamine master mix by adding 4 mL of
OptiMEM and 100 L of Lipofectamine 2000 per plate to be
transfected. Use 5% more of each reagent to allow for pipetting
error. Incubate the Lipofectamine master mix for 5 min at
room temperature.
79
4. Meanwhile, prepare the DNA master mixes: mix 15 g of BBB

decay reporter plasmid and 8 g pTracer plasmid in 4 mL
OptiMEM. Vortex.
5. Mix 4 mL of the Lipofectamine master mix slowly using stirring motions into 4 mL of the DNA master mix and incubate
at room temperature for 20 min.
6. Remove the medium from the HeLa Tet-Off cells and replace
with 8 mL of OptiMEM.
7. Add the transfection mixes from step 4 to each plate.
8. Incubate the plates with the transfection mixes for 45 h at
37C, 5% (v/v) CO2 (see Note 10).
9. Remove the transfection mix from the cells and add 18 mL of
HeLa Tet-Off medium to each plate. Let the cells recover
overnight.
10. The next day, remove the medium completely, wash cells with
PBS and remove cells from the plate with 2 mL TrypLE
express.
11. Split the cells 1:4 on p10 plates in a total of 8 mL medium the
next day. And incubate again over night at 37C, 5% (v/v)
CO2.
12. The next day, add doxycycline to a final concentration of
300 ng/mL (8 L) to each plate.
13. Incubate the plates at 37C and 5% (v/v) CO2 for 0, 1, 2, or
3 h.
14. Remove medium as completely as possible.
15. Extract total RNA with the RNeasy Mini kit following manufacturers recommendations.
16. Elute total RNA in 50 L RNase-free water.
17. Estimate RNA concentration using a spectrophotometer. The
RNA concentration should be between 1 and 3 g/L.
3.1.4. Northern Blotting
1. Dilute 10 g of total RNA into 10 L of RNase-free water.

2. Mix 10 L of NorthernMax Gly Sample Loading Dye with
the sample.
3. Spin sample for 30 s to collect all liquid at the bottom.
4. Denature the samples at 50C for 3060 min in a heat block.
5. Meanwhile, melt 1 g of agarose in 90 mL of RNase-free water;
be sure that the agarose melted completely.
6. After the agarose has cooled down to about 60C add 10 mL
of 10 NorthernMaxGly 10 Gel Prep/Running Buffer and
swirl.
80
7. Pour the gel solution into a level tray to a thickness of about

6 mm and let the gel solidify for about 30 min; the comb
should be about 1 cm from the end.
8. Prerun the gel at 85 V for about 1 min in 1 NorthernMaxGly
Gel Prep/Running Buffer prior to loading the samples.
9. Spin samples briefly to collect all liquid at the bottom prior to
loading.
10. Run the gel at 5 V/cm of distance between anode and cathode
until the blue dye front reaches the bottom of the gel.
11. Take a picture of the gel under UV-light (see Note 11).
12. Remove excessive gel by cutting the sides and top below the
slots.
13. Assemble blotting apparatus: place the gel upside down on a
mirror.
14. Prewet the membrane in NorthernMax One-Hour Transfer
Buffer and place on top of the gel and roll out trapped air
bubbles.
15. Construct a filter paper stack and revert the gel (membrane
down) onto the stack.
16. Prewet a long filter paper and put it one end on top of the gel
and the other end into a chamber containing NorthernMax
One-Hour Transfer Buffer; remove air-bubbles.
17. Blot gel for 1520 min per mm gel thickness.
18. Remove the membrane from the blotting apparatus and rinse
with 1 NorthernMaxGly Gel Prep/Running Buffer to
remove salt.
19. UV crosslink RNA to the membrane at 500 mJ/cm2 in a UV
stratalinker.
20. Wrap the membrane in plastic foil.
21. View the membrane under UV light and mark the prominent
18S and 28S rRNA bands.
22. The membrane can be frozen at this point at 20C.
23. Preheat the ULTRAhyb Ultrasensitive Hybridization Buffer
at 42C.
24. Place blot without the plastic wrap facing inward into a glass
hybridization tube and add 10 mL of preheated ULTRAhyb
buffer.
25. Prehybridize blot for at least 30 min prior to adding the
probe.
26. Add the radioactive probe (preparation see Subheading 3.1.4)
to the blot and hybridize the membrane at 42C for 34 h.
81
27. Discard the radioactive hybridization solution into the liquid

radioactive waste.
28. Wash blot twice with 15 mL NorthernMax Low Stringency
Wash Buffer for 10 min at 42C. (If precipitate has formed in
the buffer prewarm at 37C). Discard all washes into radioactive waste container.
29. Perform a high stringency wash with 15 mL NorthernMax
High Stringency Wash Buffer at 50C for 15 min. Discard
wash into radioactive waste container.
30. Wrap blot into clear plastic wrap and expose to a phosphorimager plate for at least 4 h.
31. Visualize Northern blot by scanning the plate in a
phosphorimager.
3.1.5. Probe Preparation
for Northern Blotting
Here, we describe the preparation of the GFP and -globin probes.

Other probes for detection of endogenous mRNAs can be prepared (see Note 8).
1. Mix 2040 ng of -globin or GFP PCR template with 5 L
10 PCR-buffer, 1 L 10 mM low dATP dNTP-mix, 25 pmol
of primer -globin+ or GFP+, 25 pmol of primer -globin- or
GFP-, and 5 L 32P-dATP in a final volume of 50 L water in
a PCR-tube.
2. Add 0.3 L of Taq-Polymerase and transfer the tubes into a
PCR machine.
3. Amplify the probe with the following cycles: 94C 5 min,
(94C 30 s, 48C 30 s, 72C 1 min) for 37 cycles, 72C
10 min, 4C indefinite (see Note 12).
4. The probe is purified afterward over a Microspin G-25 column following the manufacturers instruction.
5. Measure specific activity of the probe in scintillation counter.
(Add 1 L of probe to 5 mL of scintillation fluid, mix and
count).
6. Denature probe at 95C for 5 min and immediately place on
ice.
7. Add 1 106 cpm per mL hybridization buffer per probe into
the hybridization solution of the northern blot.
of RNAProtein
Interactions
As mRNA half-lives of cytokine transcripts vary with the activation

status of a T cell, so does the composition of the proteins bound to
mRNA decay elements. After stimulation of primary human
T cells, the cells can be lysed to generated cytoplasmic extracts. We
perform gentle lysis in a NP40-buffer which is compatible with
electrophoretic mobility shift assay (EMSA), UV crosslinking
assays, and the one-step affinity purification described below.
82
EMSA is an easy way to see proteinRNA interactions in

cytoplasmic lysates and is a good tool to identify a RNA-binding
protein associated with a cis-element of interest. If candidate RNAbinding proteins are known, this assay can be used to confirm
binding by simply adding a specific antibody against the protein of
interest to the reaction to determine if the antibody supershifts the
RNA-binding complex. If the proteins that associate with the
cytokine mRNA are not yet known, UV-crosslinking experiments
are able to give some information about the sizes of bound proteins and thus might lead to the identification by comparing to the
sizes of known candidate RNA-binding proteins. Another method
to identify proteins bound to the RNA decay elements is RNA
affinity purification followed by mass spectrometry. We describe a
one-step affinity purification in which biotinylated bait RNA is
incubated with the T cell lysate to allow association of the RNAbinding protein to the RNA. The proteinRNA complex is captured by streptavidin beads, washed several times and eluted. The
proteins in this complex can be analyzed by mass spectrometry.
3.2.1. Preparation
of Cytoplasmic Extracts
1. T cells are purified and stimulated as described in

Subheading 3.1.1.
2. Transfer the cells into 50 mL conical tubes and collect by centrifugation at 230 g for 8 min.
3. Pool cells of unstimulated or stimulated cells into one 50 mL
tube each (Never mix T cells from different buffy coats).
4. Bring the volume in each tube up to 50 mL with PBS.
5. Spin cells at 230 g for 8 min and remove the supernatant.
6. Repeat steps 6 and 7 once more. Remove the supernatant as
completely as possible without disturbing the pellet.
7. Estimate the pellet volume.
8. Resuspend cell pellets in 3 volumes of NP40 lysis buffer by
pipetting up and down a few times and transfer suspension to
a 1.5 mL precooled centrifugation tube.
9. Incubate on ice for 10 min.
10. Homogenize lysates with 50 strokes in a precooled dounce
homogenizer.
11. Remove nuclei and unbroken cells by centrifugation at 850 g
for 5 min at 4C.
12. Transfer supernatant to a new precooled 1.5 mL centrifugation tube. The lysates of the different buffy coats can now be
pooled.
13. Measure protein concentration at OD595 using the Biorad protein assay, following manufacturers recommendations.
83
14. Snap freeze the protein samples in liquid nitrogen and store
at 80C.
3.2.2. Electrophoretic
Mobility Shift Assay
1. To prepare cytoplasmic extract from stimulated and unstimulated

T cells follow the instructions of Subheading 3.2.1.
2. Prepare probe following the instructions of Subheading 3.2.3.
3. Prepare a native 5% polyacrylamide gel in 1 TB buffer.
4. Mix on ice in a test tube 10 g of cytoplasmic extract, 100 g
heparan sulfate, cold competitor probe at a final concentration of
10100 molar excess over probe, and 20,000100,000 cpm of
specific probe. For supershift assays add 2 L of specific or control antibody. Bring reaction to a total of 20 L with buffer RBB.
Probes and competitor RNA can be diluted in RBB buffer.
5. Incubate the reaction on ice for 30 min.
6. Add DNA loading dye and immediately load the entire sample
on a 5% native polyacrylamide gel (see Note 13).
7. Run the gel in 1 TB buffer at 100 V until the blue dye is 1 cm
from the end of the gel. Do not let the dye run out because the
probe is running in the front.
8. After the gel is finished remove one of the glass plates and
cover the gel with a Whatman filter paper. The gel will stick to
it and can be easily removed from the glass plate. Cover the gel
with plastic wrap. Do not wrap the plastic around the gel, otherwise it will not dry.
9. Dry the gel in the gel dryer at 80C for 2 h.
10. Expose the dried gel to film for approximately 248 h depending on signal strength.
3.2.3. End-Labeling
of Synthetic RNA Oligomer
Probes
1. Resuspend deprotected RNA Oligos in RNase-free water at a

concentration of 100 M. Dilute RNA oligos to a working
concentration of 1 M with RNase-free water.
2. Heat oligos to 95C for 3 min and put immediately on ice to
linearize the oligos. Linearized oligos can be stored at 20C.
3. Mix 23 L RNase-free water, 8 L 5 forward reaction buffer,
4 L 1 M linearized RNA-oligomer, 4 L 32P-ATP (4,500 Ci/
mmol), and 1 L of T4 PNK on ice.
4. Incubate the mixture at 37C for 30 min.
5. Purify sample over a Microspin G-25 column into a 1.5 mL
screw cap tube.
6. Add RNase-free water to a final volume of 200 L.
7. Extract protein and free nucleotides by adding 200 L of
phenol/chloroform/isoamyl alcohol (24:25:1, v/v/v) (see
Note 14).
84
8. Vortex for 30 s.
9. Spin sample at maximum speed for 5 min in a microfuge.
10. Transfer aqueous phase to a new tube.
11. Precipitate RNA by adding 1/10 volume of 3 M sodium acetate
and 2.5 volumes of 100% ethanol followed by an incubation of
2 h at 20C.
12. Pellet the radiolabeled RNA by centrifugation at 14,000 g for
20 min.
13. Remove the supernatant and dry the pellet in a vacufuge for
3 min.
14. Resuspend the probe in 20 L of RNase-free water.
(Add 1 L of probe to 5 mL of scintillation fluid, mix and
count).
16. Calculate the specific activity of the probe (see Note 15).
3.2.4. UV Crosslinking
Assays
1. To prepare cytoplasmic extract from stimulated and unstimulated T cells follow the instructions of Subheading 3.2.1.
2. Prepare probe following the instruction of Subheading 3.2.5.
Dilute the probe to 50,000 cpm/L with buffer RBB before
adding to reaction.
3. Mix 100 g heparan sulfate, 2 L RNase T1, and 50,000 cpm
of radiolabeled probe. Add RBB buffer to a final volume of
exactly 24 L including the amount of cytoplasmic extract.
4. Add 810 L of cytoplasmic protein to each tube, mix the
tubes, and then quick spin in a shielded microcentrifuge.
5. Incubate tubes at room temperature for 30 min behind a
shield.
6. Place the tubes on ice with the caps open and make sure that
water or ice does not get inside the tubes. Crosslink the protein to the RNA at 250 mJ/cm2 in a UV-Stratalinker.
7. Remove the samples from the stratalinker and add 24 L of 2
protein loading buffer.
8. Boil the samples for 5 min at 95C on a heat block.
9. Spin down the contents of the tube in a microfuge.
10. Subject samples to PAGE on a 12% SDS-gel; do not forget to
load a protein size marker. Do not let the dye run out of the gel.
11. After the gel is finished open the glass plates and carefully cut
of the lower part containing the blue dye.
12. Cover the gel with a Whatman filter paper. The gel will stick to
it and can be easily removed from the glass plate. Cover the gel
with plastic wrap. Do not wrap the plastic around the gel, otherwise it will not dry.
85
13. Dry the gel in the gel dryer at 80C for 2 h.

14. Expose the dried gel to film for approximately 248 h depending
on signal strength.
3.2.5. Probes for UV
Crosslinking Assays
1. Mix 6 L 5 buffer, 3 L 100 mM DTT, 6 L rNTP-mix,

1.2 L rUTP (100 mM), 2 pg template, 1 L RNAsin, 10 L
32P-UTP, 1 L T7-polymerase in a total of 30 L RNase-free
water in a PCR-tube.
2. Incubate reaction in a PCR machine for 1 h at 37C.
3. Add 1 L of DNase RQ1 to each sample and incubate for
another 15 min at 37C in a PCR machine.
4. Stop the reaction by adding 20 L RNA stop solution.
5. Purify samples over a Microspin G-25 column following
manufacturers instructions.
6. Bring volume up to 200 L with RNase-free water and transfer
to a screw cap tube.
7. Extract samples with 200 L of phenol/chloroform/isoamyl
alcohol (see Note 14).
8. Vortex sample for 1 min.
9. Spin sample at max speed for 10 min in a microfuge.
10. Transfer the aqueous phase to a new screw cap tube.
11. Add 200 L chloroform and repeat steps 810.
12. Add 1/10 volume of 3 M sodium acetate and 2.5 volumes of
100% ethanol.
13. Precipitate probes for at least 2 h at 20C.
14. Spin samples at max speed for 20 min in a microfuge.
15. Remove the supernatant and discard in radioactive waste.
16. Dry the pellet in a speedvac for 5 min.
17. Resuspend the RNA-probe in 20 L of RNase-free water.
(Add 1 L of probe to 5 mL of scintillation fluid, mix, and
count).
3.2.6. Isolation
of RNA-Binding Proteins
by One-Step Affinity
Chromatography
Note: All steps are performed at 4C or on ice.

1. Add 300 pmol of biotinylated RNA (bait), 750 pmol competitor
RNA, and 1.5 mg heparan to 10 mg of T cell lysate (total volume
about 1 mL) prepared as indicated under Subheading 3.2.1.
2. Incubate reaction at 4C for 4 h tumbling on a wheel to assure
proteinRNA interaction.
3. Meanwhile transfer 150 L of M-280 streptavidin beads per
pull down to a tube.
86
4. Wash the beads two times with 1 mL RNase-free water, once

with 1 mL RBB buffer and finally add the 50 L of RBB buffer
per reaction to the beads.
5. Stir up beads by inverting the tube a couple of times and add
50 L of M-280 streptavidin bead slurry to each tube.
6. Tumble reaction at 4C for 1 h on a wheel to allow RNA-bead
binding.
7. Collect beads on the side of the tube with a magnet.
8. Remove the supernatant.
9. Add 1 mL of cold RBB buffer to wash unbound protein off the
beads.
10. Collect beads on the side of the tube with a magnet.
11. Remove the supernatant.
12. Repeat steps 79 twice more.
13. Resuspend the beads in 20 L of RBB buffer supplemented
with 0.05% (w/v) sodium dodecylsulfate (SDS).
14. Incubate sample at 65C for 15 min to elute the protein from
the beads.
15. Collect beads using a magnet.
16. Transfer eluate to a new tube.
17. Spin sample at 7,600 g for 2 min to remove residual beads.
18. Transfer the supernatant to a new tube.
19. Freeze samples in liquid nitrogen.
20. Samples can be store at 80C prior to mass spectrometric
analysis.
4. Notes
1. In all experiments RNase-free water is required. To prepare
RNase-free water add 0.9 mL of diethylpyrocarbonate (DEPC)
to 1 L of water. Stirr overnight and autoclave for 1 h to inactivate the DEPC.
2. When using more than one buffy coat for an experiment be
sure to never mix T cells from different donors. This will lead
to activation of these cells and will compromise the
experiment.
3. Be very careful when overlaying the Ficoll-paque with blood.
Mixing of the phases results in dramatically reduced T cell
yields. Also, keep the brake of the centrifuge of. Reducing the
speed of the rotor too fast will result in mixing the phases and
an increased loss of T cells.
87
4. To stimulate T cells with anti-hCD3 antibody and antihCD28 antibody use no other plates than the ones indicated in
Subheading 2.1.1, item 2. We tested different plates and these
were the best in immobilizing the antibodies on their surface.
5. Swirl the plates well to equally distribute the Actinomycin D.
Do not add Actinomycin D to the 0 h timepoint.
6. Be sure to dry the membranes of the RNeasy column extensively. Contamination of ethanol in the RNA samples will make
the sample leak out of the pockets when northern blot is performed. Residual ethanol also has negative effect on reverse
transcription reactions.
7. T cells do not have much RNA, do not expect high yields.
8. To test the stability of specific cytokine mRNAs, specific probes
have to be generated for each transcript. For this reverse transcribe mRNA into cDNA following the instructions of the
Superscript II reverse transcriptase protocol (Invitrogen).
Generate specific primers for a 300500 nucleotide long portion of the desired cytokine. Run a standard PCR to amplify
that fragment, which needs to be gel purified subsequently.
The purified fragment can now be used to generate a radiolabeled probe as indicated in Subheading 3.1.4.
9. When using the -globin reporter based decay assay in HeLa
Tet-Off cells be sure to use Tet system approved FBS at all
times. Other types of FBS may contain tetracycline as a contaminant which will lead to irreproducible results. If the cells
are cultured with nonapproved FBS you can recover them by
growing them on a plate for a week with frequent changes of
Tet system approved FBS.
10. Do not exceed 5 h of transfection due to cytotoxicity of the
Lipofectamine 2000 reagent.
11. When performing northern blot, the gel as well as the membrane
should show the two rRNA bands. If these bands are absent it
hints either at the loss of a sample (probably due to ethanol contamination), or the degradation of the RNA sample.
12. The cycling of different PCR machines will be different which
may result in a poor radiolabeled probe. Adjust the annealing
temperatures in a nonradioactive PCR to solve that problem.
13. The DNA loading dye may interfere with proteinRNA interaction. Alternatively, the dye can be loaded into a free well or
mixed with the probe-only control.
14. When phenol extracting the probe radioactive aerosols may
develop. Always perform phenol extractions in a chemical fume
hood.
15. Calculations of the specific activity of the radiolabeled probes
for EMSA and UV-crosslinking: 4 pmol of synthetic RNA
88
were added for radiolabeling and it is anticipated that 75% of

RNA remains after purification and reconstitution. This
results in a total of (4 pmol RNA/20 L reconstitution volume) 0.75 = 0.15 pmol/L of purified RNA probe. After
counting the probe the amount of radiolabeled per picomol of
RNA can be determined. Count (cpm/L)/0.15 pmol/L
RNA = cpm/pmol. For example, the specific activity of the
probe is 200,000 cpm/pmol as measured in the scintillation
counter and 50,000 cpm will be used in the experiment: 50,0
00 0.15 pmol/L/200,000 cpm/L = 0.0375 pmol/reaction.
Acknowledgments
We thank K. Rattenbacher and I.A. Vlasova for their helpful
comments and A-B Shyu for providing -globin reporter plasmids.
This work was supported by NIH grant AI072068.
References
1. Ogilvie, R.L., Abelson, M., Hau, H.H.,
Vlasova, I., Blackshear, P.J., Bohjanen, P.R.
(2005) Tristetraprolin down-regulates IL-2
gene expression through AU-rich elementmediated mRNA decay. J. Immunol. 174:
953961.
2. Hamilton, T.A., Novotny, M., Datta, S.,
Mandal, P., Hartupee, J., Tebo, J., Li, X.
(2007) Chemokine and chemoattractant receptor expression: post-transcriptional regulation.
J. Leukoc. Biol. 82: 213219.
3. Sandler, H., Stoecklin, G. (2008) Control of
mRNA decay by phosphorylation of tristetraprolin. Biochem. Soc. Trans. 36: 491496.
4. Seko, Y., Cole, S., Kasprzak, W., Shapiro, B.A.,
Ragheb, J.A. (2006) The role of cytokine
mRNA stability in the pathogenesis of autoimmune disease. Autoimmun. Rev. 5: 299305.
5. Stoecklin, G., Tenenbaum, S.A., Mayo, T.,
Chittur, S.V., George, A.D., Baroni, T.E.,
Blackshear, P.J., Anderson, P. (2008) Genomewide analysis identifies interleukin-10 mRNA
as target of tristetraprolin. J. Biol. Chem. 283:
1168911699.
6. Lam, L.T., Pickeral, O.K., Peng, A. C.,
Rosenwald, A., Hurt, E.M., Giltnane, J.M.,
Averett, L.M., Zhao, H., Davis, R.E.,
Sathyamoorthy, M., Wahl, L.M., Harris, E.D.,
Mikovits, J.A., Monks, A.P., Hollingshead,
M.G., Sausville, E.A., Staudt, L.M. (2001)
Genomic-scale measurement of mRNA turnover and the mechanisms of action of the anti-
7.
8.
9.
10.
11.
12.
cancer drug flavopiridol. Genome Biol. 2:

RESEARCH0041.
Raghavan, A., Bohjanen, P.R. (2004)
Microarray-based analyses of mRNA decay in
the regulation of mammalian gene expression.
Brief. Funct. Genomic Proteomic 3: 112124.
Raghavan, A., Ogilvie, R.L., Reilly, C., Abelson,
M.L., Raghavan, S., Vasdewani, J., Krathwohl,
M., Bohjanen, P.R. (2002) Genome-wide analysis of mRNA decay in resting and activated
primary human T lymphocytes. Nucleic Acids
Res. 30: 55295538.
Xu, N., Chen, C.Y., Shyu, A.B. (2001) Versatile
role for hnRNP D isoforms in the differential
regulation of cytoplasmic mRNA turnover.
Mol. Cell. Biol. 21: 69606971.
Hau, H.H., Walsh, R.J., Ogilvie, R.L., Williams,
D.A., Reilly, C.S., Bohjanen, P.R. (2007)
Tristetraprolin recruits functional mRNA decay
complexes to ARE sequences. J. Cell. Biochem.
100: 14771492.
Lykke-Andersen, J., Wagner, E. (2005)
Recruitment and activation of mRNA decay
enzymes by two ARE-mediated decay activation domains in the proteins TTP and BRF-1.
Genes Dev. 19: 351361.
Stoecklin, G., Colombi, M., Raineri, I.,
Leuenberger, S., Mallaun, M., Schmidlin, M.,
Gross, B., Lu, M., Kitamura, T., Moroni, C.
(2002) Functional cloning of BRF1, a regulator of ARE-dependent mRNA turnover. Embo
J. 21: 47094718.
13. Bohjanen, P.R., Petryniak, B., June, C.H.,

Thompson, C.B., Lindsten, T. (1991) An
inducible cytoplasmic factor (AU-B) binds
selectively to AUUUA multimers in the 3
untranslated region of lymphokine mRNA.
Mol. Cell. Biol. 11: 32883295.
14. Bohjanen, P.R., Petryniak, B., June, C.H.,
Thompson, C.B., Lindsten, T. (1992) AU
RNA-binding factors differ in their binding
specificities and affinities. J. Biol. Chem. 267:
63026309.
15. Fan, X.C., Steitz, J.A. (1998) HNS, a nuclearcytoplasmic shuttling sequence in HuR. Proc.
Natl. Acad. Sci. USA 95: 1529315298.
16. Ford, L.P., Wilusz, J. (1999) An in vitro system using HeLa cytoplasmic extracts that
reproduces regulated mRNA stability. Methods
17: 2127.
17. Raghavan, A., Robison, R.L., McNabb, J.,
Miller, C.R., Williams, D.A., Bohjanen, P.R.
(2001) HuA and tristetraprolin are induced
following T cell activation and display distinct
18.
19.
20.
21.
89
but overlapping RNA binding specificities.

J. Biol. Chem. 276: 4795847965.
Shim, J., Karin, M. (2002) The control of
mRNA stability in response to extracellular
stimuli. Mol. Cells 14: 323331.
Schmidlin, M., Lu, M., Leuenberger, S.A.,
Stoecklin, G., Mallaun, M., Gross, B., Gherzi,
R., Hess, D., Hemmings, B.A., Moroni, C.
(2004) The ARE-dependent mRNA-destabilizing activity of BRF1 is regulated by protein
kinase B. Embo J. 23: 47604769.
Raghavan, A., Dhalla, M., Bakheet, T., Ogilvie,
R.L., Vlasova, I.A., Khabar, K.S., Williams,
B.R., Bohjanen, P.R. (2004) Patterns of coordinate down-regulation of ARE-containing
transcripts following immune cell activation.
Genomics 84: 10021013.
Vlasova, I.A., McNabb, J., Raghavan, A.,
Reilly, C., Williams, D.A., Bohjanen, K.A.,
Bohjanen, P.R. (2005) Coordinate stabilization of growth-regulatory transcripts in T cell
malignancies. Genomics 86: 159171.
Chapter 6
Cloning of Cytokine 3 Untranslated Regions
and Posttranscriptional Assessment
Using Cell-Based GFP Assay
Latifa Al-Haj and Khalid S.A. Khabar
Abstract
Cytokine biosynthesis is tightly regulated by a number of processes, including gene expression control.
Posttranscriptional control of cytokine gene expression offers a fine-tuning mechanism that contributes
not only to transient biosynthesis of cytokines, but also helps in rapid and early initiation of the cytokine
response. Deregulation of cytokine biosynthesis has been associated with a number of disease conditions,
including autoimmune diseases, cancer, and others. Thus, there is a need for accurate measurement of
posttranscriptional gene expression events in cytokine research. The method described here is a cell-based
GFP assay that quantitatively measures posttranscriptional effects. This method is used for assessing the
effects of modulators and conditions that lead to changes in posttranscriptional gene expression during
cytokine production or for assessment of cytokine action on posttranscriptional events of gene
expression.
Key words: Posttranscriptional regulation, 3 untranslated regions, AU-rich elements, mRNA
stability, GFP
1. Introduction
Posttranscriptional gene regulation of cytokine biosynthesis is
controlled at multiple steps, including RNA splicing, transport,
mRNA stability, and translation. The 3 untranslated region (UTR)
is a major regulatory sequence hub for posttranscriptional control,
particularly mRNA stability and miRNA-mediated regulation. Both
cis-acting and trans-acting factors can affect mRNA stability. Several
cis-acting sequence elements exist in the 3 UTR, notably, the adenylate uridylate-rich elements (ARE) which are mRNA instability
91
92
L. Al-Haj and K.S.A. Khabar
determinants; about 13% of human and mouse genes harbor AREs

in their mRNAs (1, 2). Many of the cytokine mRNAs harbor ARE
in their 3 UTR, particularly those of proinflammatory nature.
Table 1 lists examples of common cytokine ARE genes and the
sequence classes of AREs. The AREs are classified as Class I or
Table 1
Nonexhaustive list of cytokines that harbor AREs-containing 3 UTR
Definition
RefSeq
Amphiregulin
Gene
ARE
NM_001657 AREG
TAATTTATTTAAT
B-cell lymphoma/leukemia 10
NM_003921 BCL10
TTTTATTTAAATT
Bone morphogenetic protein 2
NM_001200 BMP2
ATATATTTAAAAT
NM_001202 BMP4
ATATATTTATAAC
NM_021073 BMP5
TTTTATTTATTTA
Small inducible cytokine A13
NM_005408 CCL13
TTTTATTTAAAAT
NM_004590 CCL16
TTATATTTATATT
NM_148672 CCL28
ATTTATTTATTTT
NM_002983 CCL3
TAATTTATTTATA
Small inducible cytokine A3-like 1
NM_021006 CCL3L1 TAATTTATTTATA
NM_002984 CCL4
CATTATTTATATT
NM_002984 CCL4
CATTATTTATATT
Cardiotrophin-like cytokine factor 1
NM_013246 CLCF1
ATTTATTTATTTG
Granulocyte-M colony-stimulating
factor
NM_000758 CSF2
ATTTATTTATTTATTTATTTA
Cardiotrophin-1
NM_001330 CTF1
TTTTATTTAATTT
Fractalkine
NM_002996 CX3CL1 AATTATTTATTAA
Macrophage inflammatory
protein 2-a
NM_002089 CXCL2
ATTTATTTATTTATTTATTTA
Macrophage inflammatory
protein 2b
NM_002090 CXCL3
ATTTATGTATTTA
Small inducible cytokine B5
NM_002994 CXCL5
ATATATTTATATA
Endothelin-2
NM_001956 EDN2
AATTATTTATTTT
Epiregulin
NM_001432 EREG
TTTTATTTATTTT
Growth/differentiation factor 15
NM_004864 GDF15
GTATTTATTTAAA
Growth/differentiation factor 8
NM_005259 GDF8
TTTTATTTACTTT
4
(continued)
93
Table 1
(continued)
Definition
RefSeq
Heparin-binding EGF-like growth

factor
Gene
ARE
NM_001945 HBEGF
TAATTTATTTAGT
Interferon alpha-14
NM_021057 IFNA14
ATTTATTTATTTA
Interferon alpha-16
NM_002173 IFNA16
ATTTATTTATTTA
Interferon alpha-2
NM_000605 IFNA2
TTATTTATTTAAC
Interferon alpha-21
NM_002175 IFNA21
ATTTATTTATTTA
Interferon alpha-5
NM_002169 IFNA5
ATTTATTTATTTA
Interferon alpha-6
NM_021002 IFNA6
ATTTATCTATTTA
Interferon alpha-8
NM_002170 IFNA8
ATCTATTTATTTA
Interferon beta
NM_002176 IFNB1
TTTTATTTATTTA
Interferon gamma
NM_000619 IFNG
TATTATTTATAAT
Interferon omega-1
NM_002177 IFNW1
TCATTTATTTATT
Insulin-like growth
fact 2 mRNA bind protein
NM_006546 IGF2BP1 ATTTATTAATTTA
Interleukin-10
NM_000572 IL10
CAATATTTATTAT
Interleukin-11
NM_000641 IL11
ATTTATTTATTTATTTC
Interleukin-12 subunit alpha
NM_000882 IL12A
ATTTATTTATATA
Interleukin-12 subunit beta
NM_002187 IL12B
GTTTATTTATTTATTTA
Interleukin-13
NM_002188 IL13
TCTTATTTATTAT
Interleukin-15
NM_000585 IL15
TTTAATTTATTAT
Interleukin-17A
NM_002190 IL17A
ATTTATGTATTTA
Interleukin-1 beta
NM_000576 IL1B
ATTTATTTATTTATTTG
Interleukin-1 family member 8
NM_014438 IL1F8
AATTATTTACATA
Interleukin-2
NM_000586 IL2
CTATTTATTTAAA
Interleukin-20
NM_018724 IL20
ATTTATTTTTTTA
Interleukin-22
NM_020525 IL22
ATTTATTTATAGA
Interleukin-23 subunit alpha
NM_016584 IL23A
TTGTATTTATATT
Interleukin 27
NM_145659 IL27
ATATTTATTTATT
Interleukin-28B
NM_172138 IL28B
TTTTATTTATAAA
Interleukin-3
NM_000588 IL3
ATGTATTTATTTATTTA
Interleukin-4
NM_000589 IL4
ATTTATATATTTA
3
(continued)
94
Table 1
(continued)
Definition
RefSeq
Interleukin-4
Gene
ARE
NM_000589 IL4
ATTTATATATTTA
Interleukin-5
NM_000879 IL5
GTATTTATTTAAT
Interleukin-6
NM_000600 IL6
AAATATTTATATT
Interleukin-8
NM_000584 IL8
ATTTATGTATTTATTTA
Leukemia inhibitory factor
NM_002309 LIF
AAATATTTATTTT
Lymphotoxin-alpha
NM_000595 LTA
AATTATTTATTTA
Collagenase 3
NM_002427 MMP13
ATATATTTATAAG
Oncostatin-M
NM_020530 OSM
TTATATTTATAAG
Platelet-derived growth
factor B chain
NM_002608 PDGFB
AAATTTATTTATA
Secretogranin-2
NM_003469 SCG2
TTTTATTTATAAG
TGF-beta receptor type III
NM_003243 TGFBR3 ATATATTTAATAT
Tumor necrosis factor
NM_000594 TNF
ATTTATTTATTTA
Thymic stromal lymphopoietin

isoform 1
NM_138551 TSLP
TATAATTTATATA
Vascular endothelial growth factor A
NM_003376 VEGFA
TCATTTATTTATT
Lymphotactin
NM_002995 XCL1
AATTATTTATTAT
C: Clusters according to the number of repeats (Cluster I: five or more pentamer repeats; Cluster II: three
repeats; Cluster III: four repeats; Cluster IV: two repeats, and Cluster V: one repeat in U-rich context).
Clusters IIV belong to Class II AREs while Cluster V belongs to Class I AREs
Class II based on the existence of either dispersed AUUUA

pentamers in U-rich context or repeated overlapping pentamers,
respectively (3). A bioinformatics-derived 13-nucleotide core
sequence (WWWUUAUUUAUWWW) was used to classify ARE
genes into five categories based on the number of continuous
repeats of AUUUA pentamers (Cluster IV; (4)). The Class II
cytokine mRNAs, especially those of Clusters IIII, are the most
unstable mRNAs according to this classification (5, 6).
The ARE-mRNAs are regulated by an array of RNA-binding
proteins that can selectively bind to the ARE and promote their
mRNA stability or degradation (reviewed in refs. 711). A major
RNA-binding protein that promotes mRNA stabilization is the
human antigen R (HuR) which stabilizes many cytokine AREmRNA, such as IL-8, GM-CSF, and IFN-g. Whereas, tristetraproline (TTP), KSRP, and AUF1 tend to promote the decay of
ARE-mRNA both in vitro and in vivo for cytokines, such as IL-1,
95
TNF-a, IL-3, and other Class II ARE-mRNAs (1214). Several

signaling pathways regulate mobilization and activity of the RNAbinding proteins and subsequently cytokine ARE-mRNA stability
and translation. The most notable signaling pathway that regulates
ARE-mRNA stability is mitogen-activated protein kinase p38,
and its downstream target kinase, MAPK-activated protein kinase
2 (5, 12, 15, 16).
This chapter focuses on live cell-based assay which assesses the
posttranscriptional gene regulation in cytokine studies. The assay is
GFP based and thus it can be utilized to measure the total posttranscriptional outcome, e.g., mRNA stability and translation,
using fluorescence as endpoint measurements. In this protocol, the
GFP coding region is fused with the cytokine 3 UTR that contains
mRNA turnover sequence elements, such as the AREs, and thus
the fused GFP3 UTR acts as a posttranscriptional reporter. When
compared to control stable 3 UTR that lacks the specific mRNA
sequence element in question, changes due to posttranscriptional
regulation can be monitored.
2. Materials
2.1. Cell Culture
1. Human embryonic kidney epithelial cell line HEK293 (ATCC,

Manassas, VA) and THP-1 monocytic cell line (ATCC,
Manassas, VA).
2. Dulbeccos Modified Eagles Medium (DMEM) (Invitrogen,
Carlsbad, CA) or RPMI 1640 medium (Invitrogen) supplemented with 10% (v/v) heat-inactivated (56C for 1 h) fetal
bovine serum (FBS; Invitrogen), penicillin and streptomycin
(100 units/mL penicillin, and 100 mg/mL streptomycin,
Invitrogen), and 2 mM l-Glutamine (200 mM in 0.85% (w/v)
NaCl solution).
3. TrypsinEDTA solution, sterile (200 mg/L versene EDTA).
2.2. RNA Isolation,

cDNA Synthesis,
and RT-PCR
1. TRI Reagent (SigmaAldrich).

2. Chloroform.
3. Isopropanol.
4. 70% (v/v) ethanol.
5. Deionized and autoclaved water.
6. SuperScript II RNase H reverse transcriptase enzyme
(Invitrogen), 5 first-strand buffer (250 mM TrisHCl, pH 8.3;
375 mM KCl; 15 mM MgCl2), and DTT 0.1 M (Invitrogen).
7. DNTPs mix: 100 mM dCTP, dATP, dGTP, and dTTP
(Invitrogen).
96
8. RNaseOUT (40 units/mL, Invitrogen).

9. Phosphorylated Oligo(dT) 1218 sodium salt (Amersham).
10. Custom-made (ProOligo, France) HPLC-purified oligonucleotide primers for PCR.
11. Hot start Taq DNA polymerase (Qiagen).
12. QIAquick PCR purification Kit (Qiagen).
13. QIAquick gel extraction kit (Qiagen).
2.3. Agarose Gel
Electrophoresis
1. Electrophoresis running buffer: TrisborateEDTA (5 TBE)

(Sigma), dissolve one pouch 5 concentrate (0.445 M Tris
borate, 10 mM EDTA, pH 8.3) in 1 L of deionized water.
2. Electrophoresis running buffer: TrisAcetateEDTA (10
TAE) (Sigma), dissolve one pouch 10 concentrate (0.4 M
Trisacetate, 10 mM EDTA, pH 8.3) in 1 L of deionized
water.
3. Agarose gel (Invitrogen).
4. Ethidium bromide stock solution, 10 mg/mL (Sigma).
5. 10 blue juice loading buffer: 65% (w/v) sucrose, 10 mM Tris
HCl (pH 7.5), 10 mM EDTA, 0.3% (w/v) bromophenol blue
(Invitrogen).
6. 100 bp DNA ladder (Invitrogen): Ready-to-use stock is prepared at 0.2 mg/mL in 10 mM TrisHCl (pH 7.5), 10 mM
EDTA, 0.05% (w/v) bromophenol blue, and 5% (v/v)
glycerol.
7. 1 kb DNA ladder (Invitrogen): Ready-to-use stock is prepared
at 0.2 mg/mL in 10 mM TrisHCl (pH 7.5), 10 mM EDTA,
0.05% (w/v) bromophenol blue, and 5% (v/v) glycerol.
2.4. Cloning,
Purification of 3 UTR
Reporter Constructs
1. Luria-Bertani (LB) medium (Qbiogene): One pouch that contains 10 g bacto-tryptone, 5 g bacto-yeast extract, 5 g NaCl,
pH 7.0, is dissolved in 1 L of deionized water.
2. LB agar (Qbiogene): One pouch is added to 1 L of doubledistilled water (deionized water) and sterilized by autoclaving.
Allow the medium to cool to 50C, then add ampicillin to a
final concentration of 100 mg/mL, and pour plates.
3. SOC medium (100 mL): 2.0 g bacto-tryptone, 0.5 g bactoyeast extract, 1 mL 1 M NaCl, 0.25 mL 1 M KCl, 1 mL
2 M Mg2+ stock (1 M MgCl2; 1 M MgSO4), 1 mL 2 M glucose
are brought to 100 mL with distilled water, pH 7.0.
4. Competent Escherichia coli (DH5a or other strains).
5. BamHI and XbaI (Promega).
6. Restriction enzymes (XbaI and BamHI) and enzyme reaction
buffers (Promega).
97
7. Ampicillin.
8. T4 DNA Ligase 4 U/mL supplied with T4 DNA Ligase 5
buffer containing 250 mM TrisHCl (pH 7.6), 50 mM MgCl2,
5 mM ATP, 5 mM DTT, and 25% (w/v) polyethylene glycol-8000 (Invitrogen).
9. QIA prep Spin plasmid miniprep Kit (Qiagen).
10. EGFP mammalian expression vector (Gene Therapy Systems,
Inc., San Diego, CA).
2.5. Transfection
1. Lipofectamine 2000 transfection reagent (Invitrogen).

2. Opti-MEM reduced serum medium (Invitrogen).
2.6. Additional
Materials, Equipment
1. Black opaque-walled, clear-bottom 96 well plate (Matrix).

2. ZENYTH 3100 Multimode Detectors (Anthos labtec,
Eugendorf, Austria).
3. Zeiss fluorescence microscope.
4. NanoDrop UV-Vis Spectrophotometer (Labtech International).
3. Methods
Figure 1 shows a flowchart of the main steps of cytokine 3 UTRlinked posttranscriptional assessment protocol and the scheme of
the cloning approach of cytokine 3 UTR region of interest. An
example is given with IL-8 3 UTR.
3.1. Bioinformatics
Extraction of Cytokine
3 UTR
Cytokine 3 UTR is the main regulatory region for posttranscriptional control, particularly mRNA stability. The region of the 3
UTR that harbors the cytokine mRNA destability sequence elements, such as AU rich, can be obtained from GenBank sequence
files (e.g., RefSeq mRNA database) from NCBI site (http://www.
ncbi.nlm.nih.gov). Extraction of the 3 UTR sequence can be performed manually using bioinformatics programs packages, such as
Vector NTI or LaserGen software. Use of databases, such as
UTRdb, can facilitate multiple queries search. Batch 3 UTRs can
also be executed using biomart bioinformatics data management
system (http://www.biomart.org). The identity of ARE-mRNAs
and the location of AREs can be obtained from ARED-Organism
(http://brp.kfshrc.edu.sa/ARED).
3.2. RT-PCR
of Cytokine 3 UTR:
Total RNA Extraction
and cDNA Synthesis
Complete media are prepared for each cell line. Cells are maintained in 5% (v/v) CO2 atmosphere at 37C incubator. The source
of the cytokine 3 UTR can be either genomic DNA or cDNA
synthesized from total RNA extracted from cells of interest (see
Note 1).
98

ARED-Organism; http://brp.kfshrc.edu.sa
Bioinformatics
ARE-Genes/Region
Primers/PCR
3UTR RT-PCR
IL8
102
401
IL8 CD
ARE
972
1209 pA
cDNA
BamH IXba I
Cloning
Promoter
EGFP Reporter
pA
vector
BGH 3'UTR
Control 3'UTR
BamH I
Promoter
EGFP Reporter
Xba I
pA
3'UTR
vector
Cytokine ARE 3 UTR

Cellular Models/Transfection
Endpoint Measurements
Fluorescence, RT-PCR
Fig. 1. Schematic diagram representation of the sequential steps used for reporter 3 UTR cloning. An example is given with
IL-8 3 UTR.
1. Total RNA from THP-1 monocytic cell line (see Note 2) is

extracted using TRI reagent. Cells are isolated by centrifugation and then lysed by adding 1 mL of TRI reagent per 5 106
cells. Cells are pipetted up and down to homogenize the cell
lysate, and then 200 mL of chloroform is added per 1 mL of
TRI reagent used. Samples are mixed very well and centrifuged
at 12,000 g for 15 min at 28C. After centrifugation, the
aqueous phase is transferred to a clean tube making sure not
to aspirate from the organic phase and 500 mL of isopropanol
is added per mL of TRI reagent used. Samples are allowed to
99
stand for 10 min at room temperature, and then centrifuged at

12,000 g for 15 min at 28C. The supernatant is removed
and washed once with 75% (v/v) cold ethanol, and then centrifuged at 7,500 g for 5 min at 28C. The RNA pellet is dried
for 10 min at room temperature and then appropriate volume
of nuclease-free water is added to each sample.
2. RNA concentration is calculated by measuring OD at 260 nm.
The 260/280 nm ratio is calculated to estimate the purity of
the RNA preparation; a ratio of 1.62.1 indicates pure RNA
preparation.
3. RNA samples (1 mL) are analyzed on a 1% agarose gel to confirm the quality of the RNA.
4. The cDNA synthesis is performed by standard reverse transcriptase technique. Briefly, 5 mg total RNA in total volume of
11 mL in a 1.5 mL microcentrifuge tube is incubated with
500 ng oligo(dT) at 65C for 10 min and then quick chilled
on ice for 2 min. Eight microliters of reaction mix is added to
each tube; reaction mix is prepared by adding 4 mL 5 firststrand buffer, 2 mL 0.1 M DTT, 1 mL RNaseOUT (40 units),
and 1 mL SuperScript II RT (200 units). Tubes are incubated
at 42C for 1 h, and then the reaction is inactivated by heating
at 70C for 15 min.
3.3. Primer Design
and PCR
1. The present method utilized IL-8 3 UTR as an example of

cytokine 3 UTR (see Fig. 2).
A 237-base pair region (nucleotides 9721,209 of accession number NM_000584) from the 1,250-base pair IL-8 3
3 UTR is amplified by RT-PCR (see Note 3). The forward
primer with BamHI site (underlined) is 5 CAGCAGGATCC
GATGTTGTGAGGACATGTG 3 and the reverse primer with
the XbaI site (underlined) is 5 CGACCTCTAGAACCC
TGATTGAAATTTAT 3. The additional 5 sequences are for
thermal stability of the oligonucleotides and facilitation of
restriction reaction. As a control, the eukaryotic translation
elongation factor 1 alpha 1 (EEF1A1) 3 UTR is amplified
with the following primers: the forward primer with BamHI
site (underlined) is 5 GCACCGGATCCAATATTATCCC
TAATACCTG, 3 and the reverse primer with XbaI site (underlined) is 5 GCCAGTCTAGAAATAACTTAAAACTGCCA 3.
A 205-base pair region (nucleotides 1,4511,655 accession
number NM_001402) is amplified by RT-PCR (see Note 4).
2. PCRs are carried out in a 100 mL reaction by adding 1 mL
cDNA containing 250 ng, 10 mL 10 PCR buffer, 2 mL dNTPs,
2 mL of the forward primer, 2 mL of the reverse primer, and
two units hot-start Taq DNA polymerase.
100
Fig. 2. Example of the assay performance. (a) Schematic diagram of reporter vector harboring IL-8 3 UTR containing
AU-rich element (ARE) region. The AREs are underlined. Sequence in bold is Cluster II ARE. (b) Cells were transfected with
25 ng of control (EEFA1A) and IL-8 3 UTR-fused GFP vectors. After approximately 48 h, the cells were visualized using fluorescence microscope. Captured images were analyzed using algorithm that quantitates total fluorescence intensities in
pixels. Data are presented as mean SEM of four replicate readings (c) *** denote p values < 0.001.
3. Samples are amplified by standard PCR: template denaturation

at 94C for 45 s, primer annealing at 50C for 45 s, template
elongation at 72C for 2 min, for 35 cycles. Ten-microliters
aliquot of the PCR product is ran on 1% agarose gel in 1 TBE
buffer to check the quality of the PCR product.
3.4. Cloning into EGFP
Mammalian
Expression System
1. The PCR products (IL-8 and EEF1A1 3 UTR) are purified by

a spin column (QIAquick PCR purification Kit) and eluted in
40 mL deionized water.
2. The purified PCR products are cut by BamHI and XbaI
sequentially (see Notes 5 and 6) and followed by phenol extraction and ethanol precipitation.
3. EGFP expression vector is digested with BamHI and XbaI
sequentially (see Note 7) and purified by phenol extraction and
ethanol precipitation (see Note 8).
4. The concentrations of the insert (IL-8 or EEF1A1 3 UTR)
and the vector are determined using NanoDrop UV-Vis
Spectrophotometer.
101
5. Ten microliters ligation reactions are set up as follows: The

digested and purified inserts and EGFP plasmid are mixed in
3:1 access molar ratio (see Note 9). One hundred nanogram
cut EGFP plasmid is added to 10 mL digested and purified PCR
product, 2 mL of 5 T4 DNA ligase buffer, and 1 mL T4 DNA
ligase. Ligation reaction is incubated at 16C overnight.
6. For transformation preparation: One vial containing 100 mL of
competent cells (DH5a cells) is thawed for each ligation reaction on ice and then 3 mL of ligation reaction is added (~50
150 ng). Transformation reaction is incubated on ice for
20 min and then heat shocked for 45 s at 42C and placed on
ice for 2 min.
7. SOC medium (350 mL) is added and sample is incubated at
37C for 1 h in shaking incubator.
8. Transformation reaction (100 or 200 mL) is applied onto LB
agar plate containing 100 mg/mL ampicillin. The plates are
incubated inverted at 37C incubator overnight.
9. Several white colonies are inoculated into 2 mL LB medium
and incubated at 37C overnight in a shaking incubator (see
Note 10).
10. Recombinant colonies are verified by PCR using a vector-specific forward primer and the EEF1A1 3 UTR or IL-8 3 UTR
reverse primer.
3.5. Posttranscriptional
Assessment
of Cytokine 3 UTR
Effect
1. HEK293 cells (3 104 cells per well) are grown overnight to

70% confluence in black, clear-bottomed 96-well plates.
2. The cells are transfected with 25 ng of the GFP plasmids (see
Notes 11 and 12) diluted from the stock plasmid in OptiMEM-reduced serum medium in 10 mL volume. Lipofectamine
2000 (0.5 mL of lipofectin in 10 mL Opti-MEM) is mixed gently with the DNA solution for 20 min before adding the complex onto the cells.
3. Transfection is continued for 5 h and then medium is replaced
with fresh complete medium (see Note 13).
4. All transfections are performed in quadruplicate. The variance
in GFP fluorescence among replicate microwells was 7%.
Transfection efficiency in HEK293 was always 80% 10%.
5. Day 1 and day 2 after transfection, the plates are read using
bottom-read fluorescence reader. The ZENYTH 3100 is used
with the following parameters: excitation filter (nm), 485 SL 1;
emission filter (nm), 535 SL 1; integration time (s), 1; and
bottom fluorescence read setting (see Note 14).
6. An example of the performance of the assay is given in Fig. 2.
102
4. Notes
1. The source for amplifying the cytokine 3 UTR sequence can
be the genomic DNA in case that the 3 UTR is not residing in
two exons, i.e., is not within spliceable region. If the 3 UTR
occurs in two exons, i.e., in spliced region, then cDNA is used
from total RNA that is extracted from cells that express the
cytokine of interest. Many of the cytokine 3 UTR appear to
have the 3 UTR within the whole last exon. Examples are
GM-CSF, CXCL-2, IL-1b, IL-8, and TNF-a 3 UTR.
However, the protocol that is given here is from cDNA which
is applicable for both spliced and unspliced 3 UTR.
2. The monocytic cell line (THP-1) is an excellent source of
cytokine mRNA expression when stimulated with LPS (10 mg/
mL) and cycloheximide (5 mg/mL) for 24 h (5, 6, 17).
3. The length and the region of the cytokine 3 UTR to be cloned
are subject to the experimental system in question. The whole
3 UTR can be very long and in many instances the mRNA
destability elements are clustered in specific regions. Thus, a
region of 200300 bases can be sufficient; smaller regions have
also been shown to harbor significant decay-promoting activity. In case of ARE, there may be a core ARE, usually of Class
II, and other accessory ARE. For example, 60-base region of
IL-8 contains both core and accessory ARE (18) that can
mediate the decay activity.
4. The EEF1A1 3 UTR has two polyadenylation sites giving rise
to potential alternative polyadenylation transcripts, a short 3
UTR (295 bases and transcript of 1.7 kb), and a long 3 UTR
(2121 bases and transcript of 3.5 kb). The most abundant
form is the transcript with the short 3 UTR in which 205
bases control 3 UTR was amplified (19).
5. First, digest (10 mg) PCR product with restriction enzyme that
needs the buffer with the lowest salt concentration, in this case
XbaI, and then add BamHI. If needed, the salt is removed and
the DNA is subjected to an ethanol precipitation before the
addition of the second enzyme.
6. Although in general digestion of PCR product is less efficient
than plasmids, the combination of the selected restriction enzymes,
PCR primer design, and digestion/purification protocol described
here results in efficient digestion of the PCR product.
7. We routinely use this EGFP expression vector which is under
the control of CMV IE promoter and also contains CMV
intron A. It gives strong expression which is important when
using cytokine 3 UTR that tends to destabilize the GFP mRNA
and subsequently the fluorescent protein levels. The advantage
103
of using EGFP as opposed to luciferase system is that it can be

performed on live cells with repeated measurements and does
not require cell lysis and reagent reaction leading to low
intrareplicate variance.
8. If needed to reduce clone background, dephosphorylate the
vector, and use 0.01 unit of calf intestinal alkaline phosphatase
(CIP) per each pmol of DNA ends. Molarity of ends = [(mg/mL)/
(base pairs 650 Da)] 2 ends.
9. Calculate a 1:1 molar ratio: (bp insert)/(bp vector) = (ng insert)/
(ng vector). Then, multiply by three, and use 100 ng vector.
10. Do not incubate plates for longer than 20 h to avoid the formation of satellite colonies.
11. The amount of EGFP expression vector is important to be
optimal not saturated so that the effects of reduction of the
reporter due to the cytokine 3 UTR decay-promoting action
can be appreciated. This is dependent on the cell line used, and
thus various doses of the expression vector may need to be
tested. Also, CMV promoter can be subject to transcriptional
induction in certain experimental settings, thus the control 3
UTR is always needed.
12. Prepare pure plasmid DNA for transfection; several manufacturers kits provide transfection-quality plasmid isolation kits.
13. In HEK293 and other robust cell lines, the medium containing the lipofectin may not be changed. The described method
works quite well with HEK293, Huh7, and HeLa. In the latter
case, 50 ng of the plasmid is required.
14. Fluorescence readings of lower EGFP-emitting cells, e.g., especially with certain cell lines that are hard to transfect, can be
problematic because of the low sensitivity of certain bottomread fluorimeters. An excellent alternative is the use of the modern imaging apparatus that utilize automated image focus and
capture followed by segmentation and processing algorithms.
Alternatively, pictures can be taken using standard fluorescence
microscope (optimum excitation wavelength: 488 nm and
emission wavelength: 503 nm). In all cases, exposure times and
other settings are kept constant to allow equal comparison of
experiments. The total cell and well area (pixels) and total green
intensity within the cell regions are calculated using automated
thresholding algorithms. The software tool is implemented in
MATLAB (The Mathworks, Inc., Natick, MA, USA) programming environment, and images are processed in batch mode.
15. The described assay measures the combined outcome of
posttranscriptional gene expression, including mRNA stability
and translation. The mRNA stability effects can be monitored
by GFP mRNA measurements, e.g., Northern and RT-PCR,
and use of transcription inhibitors.
104
References
1. Halees, A.S., El-Badrawi, R., Khabar, K.S.
(2008) ARED Organism: expansion of ARED
reveals AU-rich element cluster variations
between human and mouse. Nucleic Acids Res.
36: D137140.
2. Bakheet, T., Williams, B.R., Khabar, K.S.
(2006) ARED 3.0: the large and diverse
AU-rich transcriptome. Nucleic Acids Res. 34:
D111114.
3. Chen, C.Y., Shyu, A.B. (1995) AU-rich elements:
characterization and importance in mRNA degradation. Trends Biochem. Sci. 20: 465470.
4. Bakheet, T., Frevel, M., Williams, B.R.G.,
Greer, W., Khabar, K.S.A. (2001) ARED:
Human AU-rich element-containing mRNA
database reveals an unexpectedly diverse functional reportiore of encoded proteins. Nucleic
Acids Res. 29: 246254.
5. Frevel, M.A., Bakheet, T., Silva, A.M., Hissong,
J.G., Khabar, K.S., Williams, B.R. (2003) p38
Mitogen-activated protein kinase-dependent
and -independent signaling of mRNA stability
of AU-rich element-containing transcripts.
Mol. Cell. Biol. 23: 425436.
6. Raghavan, A., Dhalla, M., Bakheet, T., Ogilvie,
R.L., Vlasova, I.A., Khabar, K.S., Williams,
B.R., Bohjanen, P.R. (2004) Patterns of coordinate down-regulation of ARE-containing
transcripts following immune cell activation.
Genomics 84: 10021013.
7. Khabar, K.S. (2007) Rapid transit in the
immune cells: the role of mRNA turnover regulation. J. Leukoc. Biol. 81: 13351344.
8. Eberhardt, W., Doller, A., Akool el, S.,
Pfeilschifter, J. (2007) Modulation of mRNA
stability as a novel therapeutic approach.
Pharmacol. Ther. 114: 5673.
9. Seko, Y., Cole, S., Kasprzak, W., Shapiro, B.A.,
Ragheb, J.A. (2006) The role of cytokine
mRNA stability in the pathogenesis of autoimmune disease. Autoimmun. Rev. 5: 299305.
10. Khabar, K.S. (2005) The AU-rich transcriptome: more than interferons and cytokines, and
its role in disease. J. Interferon Cytokine Res.
25: 110.
11. Asirvatham, A.J., Magner, W.J., Tomasi, T.B.
miRNA regulation of cytokine genes. Cytokine
45: 5869.
12. Hitti, E., Iakovleva, T., Brook, M.,
Deppenmeier, S., Gruber, A.D., Radzioch, D.,
Clark, A.R., Blackshear, P.J., Kotlyarov, A.,
13.
14.
15.
16.
17.
18.
19.
Gaestel, M. (2006) Mitogen-activated protein

kinase-activated protein kinase 2 regulates
tumor necrosis factor mRNA stability and
translation mainly by altering tristetraprolin
expression, stability, and binding to adenine/
uridine-rich element. Mol. Cell. Biol. 26:
23992407.
Carballo, E., Lai, W.S., Blackshear, P.J. (2000)
Evidence that tristetraprolin is a physiological
regulator of granulocyte-macrophage colonystimulating factor messenger RNA deadenylation and stability. Blood 95: 18911899.
Winzen, R., Thakur, B.K., Dittrich-Breiholz,
O., Shah, M., Redich, N., Dhamija, S., Kracht,
M., Holtmann, H. (2007) Functional analysis
of KSRP interaction with the AU-rich element
of interleukin-8 and identification of inflammatory mRNA targets. Mol. Cell. Biol. 27:
83888400.
Winzen, R., Kracht, M., Ritter, B., Wilhelm, A.,
Chen, C.Y., Shyu, A.B., Mller, M., Gaestel, M.,
Resch, K., Holtmann, H. (1999) The p38
MAP kinase pathway signals for cytokineinduced mRNA stabilization via MAP kinaseactivated protein kinase 2 and an AU-rich
region-targeted mechanism. EMBO J. 18:
49694980.
Mahtani, K.R., Brook, M., Dean, J.L., Sully, G.,
Saklatvala, J., Clark, A.R. (2001) Mitogenactivated protein kinase p38 controls the
expression and posttranslational modification
of tristetraprolin, a regulator of tumor necrosis
factor alpha mRNA stability. Mol. Cell. Biol.
21: 64616469.
Khabar, K.S., Dhalla, M., Al-Haj, L., Bakheet, T.,
Sy, C., Naemmuddin, M. (2004) Selection of
AU-rich transiently expressed sequences: reversal of cDNA abundance. RNA 10: 747753.
Winzen, R., Gowrishankar, G., Bollig, F.,
Redich, N., Resch, K., Holtmann, H. (2004)
Distinct domains of AU-rich elements exert
different functions in mRNA destabilization
and stabilization by p38 mitogen-activated
protein kinase or HuR. Mol. Cell. Biol. 24:
48354847.
Al-Zoghaibi, F., Ashour, T., Al-Ahmadi, W.,
Abulleef, H., Demirkaya, O., Khabar, K.S.
(2007) Bioinformatics and experimental derivation of an efficient hybrid 3 untranslated
region and use in expression active linear DNA
with minimum poly(A) region. Gene 391:
130139.
Chapter 7
Integrin-Targeted Stabilized Nanoparticles for an Efficient
Delivery of siRNAs In Vitro and In Vivo
Charudharshini Srinivasan, Dan Peer, and Motomu Shimaoka
Abstract
Utilizing small interfering RNAs (siRNAs) to silence disease-associated genes holds promise as a potential
therapeutic strategy. However, the greatest challenge for RNAi remains the delivery of siRNA to target
tissues or cells. Specifically lymphocytes are difficult to transduce by conventional methods but represent
good targets for anti-inflammatory therapeutics. Integrins are an important class of cell adhesion receptors
on leukocytes. Antibodies to integrins have been used to inhibit inflammatory reactions in patients. Here,
we describe a strategy to deliver the siRNA cargo to leukocytes by stabilized nanoparticles surface-decorated
with antibodies to integrin as targeting moieties. A detailed methodology for preparation of the integrintargeted stabilized nanoparticles (I-tsNPs) and their delivery in vitro and in vivo is discussed.
Key words: Liposomes, RNAi, Leukocytes, Inflammation, Hyaluronan, Antibody, Transfection,
Systemic delivery
1. Introduction
Post-transcriptional gene silencing by RNA interference (RNAi)
has shown great potential as a therapeutic tool in targeted suppression of disease causing genes. Several siRNA delivery vectors have
been investigated for their efficiency via systemic delivery in animal
models. To mention a few, non-targeted delivery vectors such as
stable nucleic acid-lipid particles (SNALP) (1, 2), lipidoids (3),
poly D,L-lactide-co-glycolide (PLGA) microspheres (4) have been
developed for RNAi. However, cell- or tissue-specific targeted
siRNA delivery is highly desirable due to improved gene silencing
and lower undesirable side effects than those compared to nontargeted delivery (57). Some targeted siRNA delivery vectors that
are recently developed are; transferrin antibody targeted cyclodextrin
105
106
C. Srinivasan et al.
(CDP) for Ewing sarcoma in mice (8), HIV-targeted antibody

protamine conjugate for mice melanoma xenografts (9), and transferrin antibody targeted cationic liposome for pancreatic tumor
xenografts in mice (10). Most of these systems have focused on
delivering siRNA to liver and tumors in animal models. Efforts to
investigate targeted siRNA delivery to inflammatory leukocyte are
vital and hold promise for future RNAi-based therapeutics in autoimmune disease, allergy, viral infections (HIV, ebola, dengue); and
blood cancers (lymphoma, leukemia, and myeloma) (6, 1114).
But the greatest challenge is the delivery of siRNA to primary leukocytes that are resistant to conventional methods of transfection
based on cationic lipid and polymers. Although electroporation
method in in vitro conditions and hydrodynamic injection that
force siRNA into cells in vivo have shown some success (15), this
may not be feasible to use systemically due to disperse distribution
of the leukocytes within the human body. To effectively deliver
siRNAs in leukocytes, we have exploited the cell surface adhesion
molecules, integrins that mediate adhesive interactions critical for
leukocyte migration to sites of inflammation (16). In particular,
the integrins 2 and 7 are exclusively expressed on leukocytes (17,
18). With this approach, integrin-targeted stabilized nanoparticles
(I-tsNPs) was developed in our laboratory by encapsulating siRNAs within nano-sized neutral liposomes that are selectively targeted to leukocytes via surface-attached antibodies to leukocyte
integrins (1922).
In this contribution, we describe the technology for developing I-tsNPs: NPs production, surface modifications, and purification. Characterizations of the NPs for their particle size, zeta
potential, antibody-binding efficiency and siRNA entrapment is
also discussed. In vitro transfection in TK-1 cells and in vivo delivery in mice have been demonstrated to show the efficacy of the
NPs. A model siRNA Ku70, a ubiquitously expressed gene (17) is
utilized to show the gene knockdown following delivery via 7
I-tsNPs (9).
2. Materials
2.1. I-tsNP Production
1. Multilamellar liposomes (MLL): L -Phosphatidylcholine

(PC Egg, Chicken), 1,2-Dipalmitoyl-sn-Glycero-3-Phosphoethanolamine (DPPE), cholesterol (Chol) (Avanti polar lipids,
Inc., Alabaster, AL). Rotary evaporator (Buchi Corporation,
Switzerland), Thermobarrel Lipex extruderTM (Lipex
biomembranes Inc., Vancover, British Columbia, Canada) and
nucleopore membranes 0.11 m pore size (Nucleopore,
Whatman). 20 mM Hepes buffered saline, pH 7.2 (Fluka, SigmaAldrich, Saint Louis, MO), and 1 phosphate buffered saline
(PBS), pH 7.4 (Cellgro, Mediatech Inc., Manassas, VA).
Integrin-Targeted Stabilized Nanoparticles for an Efficient Delivery
107
2. Hyaluronan (HA) coated nanoliposomes: Hyaluronan (HA,

751 kDa or 850 kDa, intrinsic viscosity: 1416 dL/g, Genzyme
Corp, Cambridge, MA); 400 mM 1-(3-dimethylaminopropyl)3-ethylcarbodimide hydrochloride (EDAC, Sigma-Aldrich,
Saint Louis, MO); 100 mM N hydroxysuccinimide (NHS,
Fluka, Sigma-Aldrich, Saint Louis, MO); 0.1 M sodium acetate
buffer and 0.1 M borate buffer, pH 8.6.
3. Targeting antibody at 10 mg/mL concentration (FIB504 Ratanti mouse IgG2a against 7 integrin). 1 M ethanolamine
hydrochloride, pH 8.5.
4. Purification of I-tsNP: Size exclusion column, sepharose
CL-4B beads (Sigma-Aldrich, Saint Louis, MO).
5. Freeze drying of I-tsNPs: Alpha 12 LDplus lyophilizer
(Christ, Osterode, Germany).
of NPs
2.2.1. Particle and Zeta
Potential Analysis
Particle size and zeta potential analysis: Malvern Zetasizer nano

ZSTM (Malvern Instruments Ltd., Southborough, MA), PBS 1
buffer, pH 6.7 (with 10 mM NaCl) at 20C.
2.2.2. Binding Efficiency
FACScan flow cytometer/FACSCalibur (BD biosciences), FACS

buffer (1% (v/v) FBS and 0.01% (w/v) sodium azide). TK-1 cells
(ATCC, Manassas, VA); purified 7-I-tsNP fractions from the size
exclusion column; positive control, FIB504 Rat- anti mouse IgG2a
against 7 (10 g/mL); isotype control, purified rat IgG2a (10 g/
mL), secondary antibody FITC-Anti-Rat Ab IgG2a (1 g/mL)
(BD Pharmingen).
2.2.3. siRNA Entrapment

Efficiency in NPs
1. Ku70 siRNAs from Dharmacon (Boulder, CO). The following

four used in equimolar ratios
siRNA#1:
sense 5-GCUCUGCUCAUCAAGUGUCUGdTdT-3,
antisense 5-CAGACACUUGAUGAGCAGAGCdTdT-3
siRNA#2:
sense 5-UCCUUGACUUGAUGCACCUGAdTdT-3,
antisense 5-UCAGGUGCAUCAAGUCAAGGAdTdT-3
siRNA#3:
sense 5-ACGGAUCUGACUACUCACUCAdTdT-3,
antisense 5-UGAGUGAGUAGUCAGAUCCGUdTdT-3
siRNA#4:
sense 5-ACGAAUUCUAGAGCUUGACCAdTdT-3
antisense 5-UGGUCAAGCUCUAGAAUUCGUdTdT-3.
Alternately, pre-designed ON-TARGETplus siRNA SMARTpool, Gene ID 14375 for mouse Ku70 (Dharmacon Inc.,
Boulder, CO) can also be used.
108
2. Nuclease free water (Ambion Inc., Austin, TX).

3. Human recombinant Protamine (7,500 molecular weight,
Abnova, TaipeiCity, Taiwan), or (Spermine, 348.18 molecular
weight, Spermidine 255, Biosynth International, Inc., Itasca,
IL).
4. Quant-iT RiboGreen RNA assay kit for percent entrapment
efficiency (Molecular Probes, Invitrogen, Carlsbad, CA).
2.3. In Vitro
Transfection of siRNA
Using I-tsNPs
2.4. Ku70-siRNA
Delivery In Vivo
T cell lymphoma cell line, TK-1 cells (ATCC, Manassas, VA).
1. Mice: Wild type and 7-integrin knockout mice with C57BL/6

background (Charles River Laboratories, Wilmington, MA).
Mice should be maintained in a specific pathogen-free animal
facility.
2. 27-gauge needle with a tuberculin syringe to inject to the tail
vein of the mice.
3. Bath sonicator (Fisher Scientific) to briefly sonicate liposome
suspension before injection.
4. Isolation of splenocytes:
K10 medium:
RPMI + 10% (v/v)
FCS + supplements
70 m sieves:
Nylon sieves, BD
Falcon 352350
2% (v/v) FCS:
HBSS + 2% (v/v) FCS
Frosted glass slides:

VWR 48312002
RBC (red blood cell) lysis buffer:

8.3 g/L NaCl
0.001 M TrisHCl, pH 7.5
Small Petri dishes:

BD Falcon 35 3002
3. Methods
3.1. I-tsNP Production
and Purification
I-tsNPs are nanometer sized hyaluronan coated neutral liposomes

possessing targeting moieties on their surface (antibodies to integrin molecules on leukocytes). The preparation involves two critical
processes (1) preparation of stabilized NPs by chemical conjugation of hyaluronan that coats the surface of liposomes and (2)
introduction of targeting molecules (mAbs) on the surface of the
stabilized NPs (Refer to Fig. 1, Schematic).
1. Prepare multilamellar liposomes (MLL), composed of phosphatidylcholine (PC), dipalmitoylphosphatidylethanolamine
(DPPE), and cholesterol (Chol) at molar ratios of 3:1:1
(PC:DPPE:Chol), using conventional lipid-film hydration
method (17, 18).
109
Fig. 1. The schematic showing steps involved in the production of I-tsNPs. Multilamellar vesicle (MLV) prepared by rotary
evaporation method is extruded to form a unilamellar vesicle (ULV). Hyaluronan is coated onto surface of liposomes by
covalently binding to DPPE part of the lipids in the ULV. A monoclonal antibody (mAb against integrin) is then covalently
attached to hyaluronan via an amide bond linkage forming I-tsNPs (e.g., 7 I-tsNPs). Protamine condensed siRNAs are then
entrapped within the lyophilized NPs by rehydration to form a transfection complex.
Dissolve the lipids at final concentration of 40 mg/mL in

ethanol (96%) by stirring for 3045 min at 60C. This is followed by rotary evaporation for 12 h at 65C (see Note 1).
2. Hydrate the lipid film with 20 mL of 20 mM Hepes-buffered
saline pH 7.4 or 1 PBS, pH 7.4 to create MLL. Thoroughly
Vortex until a thin milky liposome suspension is formed.
3. Incubate the liposome suspension in a shaker (~200 rpm) at
37C for 2 h to ensure complete mixing and homogeneity (see
Note 2).
4. Extrude the resulting MLL into unilamellar nano-liposomes
(ULNL) with a Thermobarrel Lipex extruder at 65C under
nitrogen pressures of 300550 psi.
5. Carry out the extrusion in a stepwise manner using progressively decreasing pore-sized membranes (from 1, 0.8, 0.6, 0.4,
0.2, to 0.1 m), with 10 cycles per pore-size.
6. ULNL are surface-modified with high molecular weight
hyaluronan (HA) (751 kDa or 850 kDa intrinsic viscosity:
1416 dL/g) as described below.
7. Dissolve 20 mg HA in 0.1 M sodium acetate buffer. Stir at
37C for 30 min to fully dissolve HA. Pre-activate with 400 mg
of EDAC, at pH 4.0 and stir for 2 h at 37C. Centrifuge the
extruded liposome suspension (ULNL) for 13 h in an ultracentrifuge and resuspend the pellet in 0.1 M borate buffer, pH
8.6. Combine the activated HA with the liposome suspension
(ULNL) in a 1:1 volume ratio and incubate overnight at 37C,
with gentle stirring. Separate the resulting HA-ULNL from
free HA by washing three times by ultra-centrifugation
(1.3 105 g, 4C, for 13 h for each wash).
8. Perform the coupling reaction of HA-modified liposomes to
mAbs using an amine-coupling method. Incubate 50 L
HA-modified liposomes with 200 L of 400 mM EDAC and
200 L of 100 mM NHS for 20 min at room temperature with
gentle stirring.
110
9. Mix the EDAC-NHS-activated HA-nano-liposomes with

50 L mAb (~500 g of FIB504 Rat-anti mouse IgG2a against
7 integrin, 10 mg/mL in HBS or PBS, pH 7.4) and incubate
for 150 min at room temperature with gentle stirring. Add
20 L 1 M ethanolamine HCl (pH 8.5) to block the reactive
residues (see Note 3).
10. Purify I-tsNPs (e.g., 7I-tsNPs) to remove uncoupled mAbs
using a size exclusion column packed with sepharose CL-4B
beads (Sigma-Aldrich, Saint Louis, MO) and equilibrated with
HBS, pH 7.4.
11. Prepare the purified particle suspensions for lyophilization:
Snap freeze 0.2 mL aliquots in a mixture of 100% (v/v) ethanol
and dry ice for ~2030 min; freeze the aliquots for 24 h
at 80C and lyophilize for 48 h using an alpha 12 LDplus
lyophilizer (see Note 4).
12. Store the lyophilized particles at 80C until further use.
3.2. I-tsNP
Characterization
3.2.1. Particle and Zeta
Potential Analysis
3.2.2. Binding Efficiency
of NPs
Measure the nanoparticle diameter and surface charge (zeta potential) using a Malvern Zetasizer nano ZS. Examples of particle size
and surface charge of the NPs are given in Table 1.
Flow cytometry is used to confirm intact binding ability of the

surface-attached antibodies.
1. Add cells (TK-1) at 0.5 106 cells/FACS tube.
2. Wash cells in 1 mL FACS buffer and centrifuge at 200 g,
5 min, 4C.
3. Resuspend the pellet with following controls and samples to a
final antibody concentration of 10 g/mL in respective tubes
in 50 L FACS buffer: (a) Cells alone as a mock control;
(b) isotype control, purified rat IgG2a; (c) Positive Control,
FIB504 Rat-anti mouse IgG2a against 7; and (d) Samples,
purified 7-I-tsNP fractions.
Table 1
Particle size and zeta potential measurements
Particle
Diameter
Zeta potential
IgG sNP
127 13 nm
18.5 1.2 mV
7 I-tsNP
139 21 nm
23.7 2.6 mV
All measurements were done in 1 PBS, pH 6.7 (with 10 mM NaCl) at

20C in a Zetasizer nano ZS, Malvern. Data presented as an average SD
from n = 4 independent experiments
111
7 I-tsNP
Cell Number
IgG sNP
101
102
103
7 expression
Fig. 2. Binding efficiency of NPs by FACS. Comparison of control nanoparticles (isotype IgG
NPs) and integrin-targeted NPs (7 I-tsNPs) shows higher binding of 7 I-tsNPs in TK-1
cells that express high levels of 7 integrin. This demonstrates the specificity of the I-tsNPs
to integrins expressed on leukocytes for targeted delivery.
5. Wash cells with 1 mL FACS buffer, centrifuge at 200 g,

5 min.
6. Resuspend the cell pellets and stain with secondary antibody
FITC-Anti-Rat Ab IgG2a (1 g/mL) in 50 L FACS buffer.
8. Wash with 1 mL FACS buffer, centrifuge at 200 g, 5 min,
4C.
9. Resuspend the cell pellets in appropriate volume of FACS buffer and analyze by FACS.
An example of FACS analysis is shown in Fig. 2.
3.2.3. Entrapment
Efficiency of NPs
Ku70 siRNAs entrapment in I-tsNPs is described as follows:

1. Mix siRNAs with full-length recombinant protamine (1:5,
siRNA:protamine molar ratio) or spermine (1:22, siRNA: spermine molar ratio) or spermidine (1:30, siRNA:spermidine
molar ratio), in nuclease free water and incubate for 20 min at
RT to form a complex.
2. For siRNA entrapment in I-tsNPs, rehydrate the lyophilized
nanoparticles (i.e., 7 I-tsNP, IgG-sNP, or sNP; 12.5 mg total
lipids for in vivo experiments and 10100 g total lipids for
in vitro experiments) by adding 0.2 mL nuclease free water
containing protamine- (or spermine) condensed siRNAs
(1,0003,500 pmol for in vivo experiments and 50750 pmol
for in vitro experiments).
112
Table 2
Number of mAb immobilized on the surface and entrapment
of siRNAs molecules
siRNAs entrapmenta
(# of molecules)
Encapsulation efficiency
of condensed siRNA
Nanoparticles type
Mean SEM
Mean SEM
IgG sNP
3,750 1,300
78 10
7 I-tsNP
4,000 1,200
80 12
The amount of siRNAs that was used for encapsulation was known. Upon
encapsulation, a RiboGreen assay (molecular probes) was preformed to assess
the amount of siRNAs that was entrapped
a
3. Perform the entrapment procedure immediately before use in

in vitro transfection or in vivo injection.
4. The concentrations of siRNAs and percent entrapment are
determined by a Quant-iT RiboGreen RNA assay
(Molecular Probes, Invitrogen, Carlsbad, CA). An example is
given in Table 2.
3.3. Ku70 siRNA
Delivery In Vitro
in TK-1 Cells
(Studied by Flow
Cytometry)
1. Plate TK-1 cells in microtiter plates (24 well plate) and culture
them overnight at 37C, 5% (v/v) CO2 (2.5 105 cells in 400 l
media/well) without serum or antibiotics.
2. Add 50 l/well of 7 I-tsNP entrapping siRNAs (e.g., Ku70siRNA) dropwise and shake gently. Spin down the plate at
300 g for 5 min. Appropriate controls should be included:
cells with no treatment; cells with Ku70-siRNA alone; cells
with negative control siRNA (e.g., silencer firefly Luciferase
siRNA or scrambled siRNA). Culture the cells for 5 h.
3. Add 50 L of serum containing culture media (10% (v/v) FBS
in RPMI) and shake gently by rocking the plate from side to
side.
4. Culture cells for further 6072 h at 37C, 5% (v/v) CO2 and
perform intracellular staining (as described below) for detection of Ku70 to confirm the effect of siRNA delivered using
I-tsNP.
3.3.1. Intracellular Staining

and Flow Cytometry
1. Transfer TK-1 cells to 96 well V bottom plates.

2. For intracellular staining cells, fix and permeabilize cells with
Fix-and-Perm KitTM (Caltag Laboratories, Burlingame, CA).
Isotype control
KD1
NE
Counts
KD2
113
100
101
102
Ku70
103
NE -Ku70 expression inTK-1 cells - mock treated cells.

KD1 - Knockdown of Ku70 using 100pmol-siRNA delivered via7 I-tsNP inTK-1 cells
KD2 - Knockdown of Ku70 using750pmol-siRNA delivered via7 I-tsNP inTK-1 cells
Fig. 3. 7 I-tsNP entrapping Ku70 siRNAs induces silencing in TK-1 cells. FACS histograms show gene silencing effect of
7 I-tsNPs entrapping different concentrations of Ku70 siRNAs (100 and 750 pmol, KD1 and KD2, respectively) in comparison to isotype NPs (gray area under the curve) and mock treated cells (NE).
3. Detection of Ku70 expression performed by adding antibody

to Ku70 (purified mouse anti-Ku70, Santa Cruz Biotechnology,
Santa Cruz, CA) at a final concentration of 10 g/mL, 50 L
in FACS buffer.
4. Incubate on ice for 30 min and counter stain with FITCconjugated Goat anti-mouse IgG (BD Pharmingen).
5. Wash with FACS buffer and perform FACS analysis for Ku70
expression.
An example of the intracellular stain is given in Fig. 3.
3.4. Delivery
of Ku70-siRNA In Vivo
1. Make groups of mice (at least 3 mice/group, preferentially

58 mice/group). Make sure to include a mock treated group.
2. Pre-heat mice with a lamp in order to expose their tail veins.
3. Prior to injection sonicate the suspension for 10 min in a
bath sonicator to dissolve any potential aggregates.
4. Use a 30-gauge needle with a tuberculin syringe to inject to
the tail vein of the mice. 100200 L/mouse with 2.5 mg/kg
(50 g) siRNA entrapped in 250 g liposomes.
5. 48 or 72 h post injection sacrifice the mice and isolate the
spleen (see Note 5).
6. Make a single cell suspension from the spleen as reported in
Note 5, and perform an intracellular staining with anti-Ku70
mAb (Santa Cruz) as detailed above.
114
4. Notes
1. Liposome preparation: Following the evaporation of organic
solvents, it is advisable to pass any inert gas like argon for
23 min to completely remove traces of organic solvent and
prevent oxidation of lipids.
2. The liposome suspension following 2 h incubation can be
stored at 4C until extrusion procedure. But prior to extrusion, the liposomes should be pre-warmed to 37C to enable
easy extrusion process.
3. Coupling reaction: EDAC/NHS-activated HA-nanoliposomes with antibody reaction mixture can be incubated
overnight at room temperature and then blocked with 20 L
of 1 M ethanolamine, pH 8.5.
4. Lyophilization of NPs: Prior to lyophilization of purified liposome fractions, the particle suspensions should be tested for
antibody-binding efficiency by FACS analysis as described in
Subheading 3.2.2. Select the fractions that give high binding
efficiency and pool all the fractions. Aliquots (0.2 mL) of the
I-tsNP suspension are added into amber glass vials prior to
lyophilization. Depending on the cells that are transfected,
I-tsNP fractions can be diluted before aliquots are prepared for
lyophilization to give an optimal transfection or gene-silencing
efficiency.
5. Isolation of splenocytes:
One spleen yields approximately 108 splenocytes, of which
~10% are CD8+ and ~20% are CD4+.
(a) Harvest spleens into K10 media, removing as much connective tissue as possible.
(b) Place ~3 mL K10 media and the splenocytes in a small
petri dish. Homogenize into a single cell suspension:
Using the flat top of a 5 mL syringe shaft, homogenize

the splenocytes
(or)
Using the frosted ends of two glass sides, homogenize

the splenocytes.
(c) Rinse the sieve or slides with K10 media or 2% (v/v) FCS,
then transfer splenocytes into a 15-mL conical tube.
(d) Spin down cells for 5 min at 320 g.
(e) Aspirate off supernatant and flick cell pellet to loosen.
(f) Resuspend cells in 2 mL RBC lysis buffer.
(g) Incubate at 37C for 5 min.
(h) Add 10 mL 2% (v/v) FCS and spin 5 min, 320 g.
115
(i) Aspirate off supernatant and flick cell pellet to loosen.

Resuspend cells in 1 mL 2% (v/v) FCS.
(j) Place 70-m sieve onto a 50-mL conical tube. Rinse the
sieve with 2% (v/v) FCS. Pass splenocytes through sieve,
rinsing with 30 mL 2% (v/v) FCS. Take an aliquot to count
cells, taking note of final volume.
(k) Spin 5 min at 200 g, aspirate off supernatant and flick cell
pellet to loosen. Resuspend cells at desired concentration.
(l) Keep cells on ice or at 4C. Cells can be kept overnight at
4C for uses such as feeder cells.
References
1. Morrissey, D. V., Lockridge, J. A., Shaw, L.,
Blanchard, K., Jensen, K., Breen, W.,
Hartsough, K., Machemer, L., Radka, S.,
Jadhav, V., Vaish, N., Zinnen, S., Vargeese, C.,
Bowman, K., Shaffer, C. S., Jeffs, L. B., Judge,
A., MacLachlan, I., Polisky, B. (2005) Potent
and persistent in vivo anti-HBV activity of
chemically modified siRNAs. Nat. Biotechnol.
23: 10021007.
2. Zimmermann, T. S., Lee, A. C., Akinc, A.,
Bramlage, B., Bumcrot, D., Fedoruk, M. N.,
Harborth, J., Heyes, J. A., Jeffs, L. B., John,
M., Judge, A. D., Lam, K., McClintock, K.,
Nechev, L. V., Palmer, L. R., Racie, T., Rohl,
I., Seiffert, S., Shanmugam, S., Sood, V.,
Soutschek, J., Toudjarska, I., Wheat, A. J.,
Yaworski, E., Zedalis, W., Koteliansky, V.,
Manoharan, M., Vornlocher, H. P., MacLachlan,
I. (2006) RNAi-mediated gene silencing in
non-human primates. Nature 441: 111114.
3. Akinc, A., Zumbuehl, A., Goldberg, M.,
Leshchiner, E. S., Busini, V., Hossain, N.,
Bacallado, S. A., Nguyen, D. N., Fuller, J.,
Alvarez, R., Borodovsky, A., Borland, T.,
Constien, R., de Fougerolles, A., Dorkin, J. R.,
Narayanannair Jayaprakash, K., Jayaraman, M.,
John, M., Koteliansky, V., Manoharan, M.,
Nechev, L., Qin, J., Racie, T., Raitcheva, D.,
Rajeev, K. G., Sah, D. W., Soutschek, J.,
Toudjarska, I., Vornlocher, H. P., Zimmermann,
T. S., Langer, R., Anderson, D. G. (2008) A
combinatorial library of lipid-like materials for
delivery of RNAi therapeutics. Nat. Biotechnol.
26: 561569.
4. Khan, A., Benboubetra, M., Sayyed, P. Z., Ng,
K. W., Fox, S., Beck, G., Benter, I. F., Akhtar,
S. (2004) Sustained polymeric delivery of gene
silencing antisense ODNs, siRNA, DNAzymes
and ribozymes: in vitro and in vivo studies.
J. Drug Target. 12: 393404.
5. Kawakami, S., Hashida, M. (2007) Targeted

delivery systems of small interfering RNA by
systemic
administration.
Drug
Metab.
Pharmacokinet. 22: 142151.
6. Dykxhoorn, D. M., Lieberman, J. (2005) The
silent revolution: RNA interference as basic
biology, research tool, and therapeutic. Annu.
Rev. Med. 56: 401423.
7. Akhtar, S., Benter, I. F. (2007) Nonviral delivery of synthetic siRNAs in vivo. J. Clin. Invest.
117: 36233632.
8. Hu-Lieskovan, S., Heidel, J. D., Bartlett, D.
W., Davis, M. E., Triche, T. J. (2005) Sequencespecific knockdown of EWS-FLI1 by targeted,
nonviral delivery of small interfering RNA
inhibits tumor growth in a murine model of
metastatic Ewings sarcoma. Cancer Res. 65:
89848992.
9. Song, E., Zhu, P., Lee, S. K., Chowdhury, D.,
Kussman, S., Dykxhoorn, D. M., Feng, Y.,
Palliser, D., Weiner, D. B., Shankar, P., Marasco,
W. A., Lieberman, J. (2005) Antibody mediated in vivo delivery of small interfering RNAs
via cell-surface receptors. Nat. Biotechnol. 23:
709717.
10. Pirollo, K. F., Rait, A., Zhou, Q., Hwang, S.
H., Dagata, J. A., Zon, G., Hogrefe, R. I.,
Palchik, G., Chang, E. H. (2007) Materializing
the potential of small interfering RNA via a
tumor-targeting nanodelivery system. Cancer
Res. 67: 29382943.
11. Behlke, M. A. (2006) Progress towards in vivo
use of siRNAs. Mol. Ther. 13: 644670.
12. Goffinet, C., Keppler, O. T. (2006) Efficient
nonviral gene delivery into primary lymphocytes
from rats and mice. Faseb J. 20: 500502.
13. Marodon, G., Mouly, E., Blair, E. J., Frisen,
C., Lemoine, F. M., Klatzmann, D. (2003)
Specific transgene expression in human and
116
14.
15.
16.
17.
18.
mouse CD4+ cells using lentiviral vectors with
regulatory sequences from the CD4 gene.
Blood 101: 34163423.
Zhang, Y., Lu, H., LiWang, P., Sili, U.,
Templeton, N. S. (2003) Optimization of gene
expression in nonactivated circulating lymphocytes. Mol. Ther. 8: 629636.
Lai, W., Chang, C. H., Farber, D. L. (2003)
Gene transfection and expression in resting and
activated murine CD4 T cell subsets. J.
Immunol. Methods 282: 93102.
Song, E., Lee, S. K., Wang, J., Ince, N.,
Ouyang, N., Min, J., Chen, J., Shankar, P.,
Lieberman, J. (2003) RNA interference targeting Fas protects mice from fulminant hepatitis.
Nat. Med. 9: 347351.
Peer, D., Park, E. J., Morishita, Y., Carman, C.
V., Shimaoka, M. (2008) Systemic leukocytedirected siRNA delivery revealing cyclin D1 as
an anti-inflammatory target. Science 319:
627630.
Peer, D., Zhu, P., Carman, C. V., Lieberman,
J., Shimaoka, M. (2007) Selective gene silencing
19.
20.
21.
22.
in activated leukocytes by targeting siRNAs to

the integrin lymphocyte function-associated
antigen-1. Proc. Natl. Acad. Sci. USA 104:
40954100.
Luo, B. H., Carman, C. V., Springer, T. A.
(2007) Structural basis of integrin regulation
and signaling. Annu. Rev. Immunol. 25:
619647.
Zhu, J., Carman, C. V., Kim, M., Shimaoka,
M., Springer, T. A., Luo, B. H. (2007)
Requirement of alpha and beta subunit transmembrane helix separation for integrin outsidein signaling. Blood 110: 24752483.
Peer, D., Shimaoka, M. (2009) Systemic siRNA
delivery to leukocyte-implicated diseases. Cell
Cycle 8: 853859.
Kim, S. S., Peer, D., Kumar, P., Subramanya,
S., Wu, H., Asthana, D., Habiro, K., Yang, Y.
G., Manjunath, N., Shimaoka, M., Shankar, P.
(2010) RNAi-mediated CCR5 silencing by
LFA-1-targeted nanoparticles prevents HIV
infection in BLT mice Mol. Ther. 18:
370376.
Chapter 8
Hammerhead Ribozyme-Mediated Knockdown of mRNA for
Fibrotic Growth Factors: Transforming Growth Factor-Beta 1
and Connective Tissue Growth Factor
Paulette M. Robinson, Timothy D. Blalock, Rong Yuan,
Alfred S. Lewin, and Gregory S. Schultz
Abstract
Excessive scarring (fibrosis) is a major cause of pathologies in multiple tissues, including lung, liver, kidney,
heart, cornea, and skin. The transforming growth factor-b (TGF-b) system has been shown to play a key
role in regulating the formation of scar tissue throughout the body. Furthermore, connective tissue growth
factor (CTGF) has been shown to mediate most of the fibrotic actions of TGF-b, including stimulation of
synthesis of extracellular matrix and differentiation of fibroblasts into myofibroblasts. Currently, no
approved drugs selectively and specifically regulate scar formation. Thus, there is a need for a drug that
selectively targets the TGF-b cascade at the molecular level and has minimal off-target side effects. This
chapter focuses on the design of hammerhead ribozymes, measurement of kinetic activity, and assessment
of knockdown mRNAs of TGF-b and CTGF in cell cultures.
Key words: Ribozymes, TGF-b, CTGF, Scar formation, Transduction, Oligonucleotides
1. Introduction
Transforming growth factor-beta (TGF-b) and connective tissue
growth factor (CTGF) play key roles in regulating scar formation
in normal wound healing in tissues throughout the body (1, 2).
Molecular analyses of pathological scars have found prolonged
elevated levels of TGF-b and CTGF mRNAs and proteins, which
has led to the hypothesis that fibrotic scars are a result of excessive
activities of these two growth factor systems. In addition, CTGF
has recently been shown to mediate most of the fibrotic activity of
TGF-b, including stimulation of synthesis of extracellular matrix
and differentiation of fibroblasts into myofibroblasts (3).
117
118
P.M. Robinson et al.
Currently, no approved drugs selectively and specifically regulate

synthesis and/or action of TGF-b and CTGF systems. Neutralizing
antibodies to TGF-b and inhibitors of TGF-b receptor kinase have
been developed and evaluated in clinical trials, but none have
received clearance from the Food and Drug Administration. Thus,
there is a need for a drug that selectively targets the TGF-b cascade
at the molecular level that produces minimal off-target side effects.
Our approach has been to develop gene-specific, oligonucleotidebased drugs, specifically hammerhead ribozymes that cleave TGF-b
and CTGF mRNA molecules.
This chapter gives a start to finish description of how to design,
validate, and test in cell culture hammerhead ribozymes that target
specific genes and could potentially be used as a therapeutic agent.
The first method describes the general steps to design a hammerhead
ribozyme. This method is the same for the design of all hammerhead ribozymes. The second method describes how to validate a
hammerhead ribozymes ability to cleave its targeted mRNA
substrate. The ribozyme cleavage time course study and the
ribozyme multiturnover study can be done with minimal variation
to assess the activity of different ribozymes. Two general approaches
are typically used to deliver ribozymes to cells in culture: direct
delivery of a chemically modified preformed ribozyme that resists
enzymatic degradation by RNases and stable transfection with a
plasmid expressing the ribozyme (4). Unprotected, nonchemically
modified ribozymes have a half-life of seconds in the body and cell
culture. Chemically protected ribozymes can be stabilized with a
half-life that can be measured for several hours (5). To circumvent
the short half-life of a ribozyme, construction of a plasmid that
expresses the ribozymes allows for constitutive expression of the
ribozyme. The third method describes the techniques used to make
the plasmids containing either TGF-b1 ribozyme or the CTGF
ribozyme. The last three sections describe methods to test the
efficiency of ribozymes in cell culture using different transfection
techniques and four different ways to determine knockdown of
the selected protein.
2. Materials
2.1. Ribozyme Design
and Synthesis
2.2. Ribozyme
Time-Course and
Multiturnover Kinetics
Mfold program:
mfold.html.
http://bioweb.pasteur.fr/seqanal/interfaces/
1. Oligo deprotection and labeling with g-[32P]-dATP: OligoRNA (10 pmol/mL), 1 mL RNasin (Promega; Madison, WI,
USA), 1 mL 0.1 M dithiothreitol (DTT), 3 mL double-distilled
water (ddH2O), 1 mL [g32P]-dATP, 1 mL 10 PNK buffer, and
1 mL T4 polynucleotide kinase (Roche Molecular Biochemicals;
Indianapolis, IN, USA).
Hammerhead Ribozyme-Mediated Knockdown of mRNA
119
2. Phenol/chloroform/isoamyl alcohol extraction: A solution of

the ratio of 25:24:1, respectively.
3. Sephadex G25 fine spin column (Roche Applied Science).
4. Ribozyme cleavage buffer: 40 mM Tris/HCl, pH 7.5, and
20 mM MgCl2.
5. Ribozyme cleavage stop solution: 6 mL of 90% formamide,
50 mM EDTA (pH 8.0), 0.05% xylene cyanol, and 0.05%
bromophenol blue.
6. Polyacrylamide urea running buffer: 1 Tris/borate/EDTA
(TBE) (obtained by adding 200 mL of 5 Novex TBE
Running Buffer to 800 mL of deionized water).
2.3. Plasmid
Construction
1. Single-stranded synthetic DNA oligonucleotides encoding

complementary sequences.
2. Restriction enzymes: NsiI and HindIII.
3. Initial plasmid construct: pTR-UF21HP.
2.4. Analysis of
Endogenous Target
mRNA Knockdown
by a Ribozyme
2.4.1. CTGF Ribozyme
Analysis
1. Dulbeccos modified Eagles medium (DMEM), Medium 199,

Hams F12 nutrient mixture containing 1 mM NaHCO3, and
buffered with 25 mM HEPES at pH 7.4. The medium is supplemented with 10% heat-inactivated normal calf serum and
1 antibiotic-antimycotic (Gibco BRL).
2. 200 mg/mL geneticin (G418 Sulfate) dissolved in cell culture
media.
Human Cell Culture

and Transfection
Quantitative Reverse
Transcription-Polymerase
Chain Reaction
1. TRIzol reagent (Invitrogen, Gaithersburg, MD).

2. 1 TaqMan One-step reverse transcription-polymerase chain
reaction (RT-PCR) Master Mix, 900 nM forward, 900 nM
reverse primer, 2 mM fluorescent TaqMan probe, and RNA
sample (CTGF mRNA standard or 500 ng of sample RNA) to
a final volume of 25 mL per reaction.
3. TaqMan glyceraldehyde phosphate dehydrogenase (GAPDH)
Control Kit (Applied Biosytems, Foster City, CA, USA).
CTGF Enzyme-Linked
Immunosorbent Assay
1. Biotinylated and nonbiotinylated, affinity-purified goat polyclonal antibodies.

2. Blocking buffer: Phosphate-buffered saline (PBS)/0.02%
sodium azide/1% bovine serum albumin.
3. Alkaline phosphatase-conjugated streptavidin (1.5 mg/mL,
Zymed, South San Francisco, CA, USA).
4. Alkaline phosphatase substrate solution (1 mg/mL p-nitrophenyl phosphate).
120
5. Sodium carbonate/bicarbonate buffer/0.02% sodium azide,

pH 9.6.
2.4.2. TGF-b1 Ribozyme
Analysis
Mouse-Immortalized Cell
Culture and Transfection
1. Equal parts Hams F-12, Medium 199, and DMEM media,

containing 20 mM HEPES, 1 mM NaHCO3, 100 U/mL
penicillin, and 100 mg/mL streptomycin, supplemented with
10% normal goat serum.
2. Hypoosmolar electroporation buffer (Eppendorf Scientific,
Inc., Germany).
3. 200 mg/mL geneticin (G418 Sulfate) dissolved in cell culture
media.
RNA Extraction and

Reverse TranscriptionPolymerase Chain Reaction
1. TRIzol reagent.
2.5. Analysis of
Exogenous Synthetic
Target Knockdown by
a Ribozyme (TGF-b1):
Human Embryonic
Kidney298 Cell Culture
and Dual Transfection
1. DMEM, with 4.5 g/L glucose and 1 g/L L-glutamine.
2. Superscript TM First-strand synthesis system for RT-PCR

(GIBCO BRL).
2. Turbofect reagent (Fermentas Inc.; Glen Burnie, MD, USA).

3. Quanti-BlueTM (InvivoGen, San Diego, CA, USA).
3. Methods
3.1. Ribozyme Design
and Synthesis
All potential ribozyme cleavage sites within the human CTGF or

TGF-b1 cDNA sequences were initially identified based on the
optimal G-U-C nucleotide sequence for hammerhead ribozymes (6).
Potential cleavage sites were then evaluated for secondary folding
structures around the G-U-C sequence using the theoretical lowest energy conformations calculated using the Mfold program
(http://bioweb.pasteur.fr/seqanal/interfaces/mfold.html) (7). Only
those sites for which the G-U-C sequence was in a single-stranded,
nonbase-paired region were considered further. In addition, the
nucleotide sequence of the 20mer centered on the G-U-C site
were examined for the relative content of A and U bases, since
previous studies had shown that flanking sequences with higher
numbers of A and U bases tend to have more rapid release rate
constants compared to flanking sequences that are rich in G-C
content. The lengths of the 5 and 3 hybridization arms, which
comprise the helical stems I and III of the ribozymes, were five
and six nucleotides, respectively. The catalytic core structure was
formed by the 21mer with the nucleotide sequence of
CUGAUGAGGUCCUUCGGGACGAA (Fig. 1). Taken together,
121
Fig.1. Sequence and secondary structure of the synthetic RNAs and their targets. The
uppercase letters represent the ribozyme RNA sequences, and the lowercase letters
represent the target RNA sequences. Roman numerals label the helices. Arrows indicate
the site of c1eavage. (a) CTGF hammerhead ribozymes targeting nucleotide sequences at
positions CHR 745 and CHR 859, (b) TGF-b1 hammerhead ribozyme targeting nucleotide
sequences at positions THR 576 and THR 1429.
the hammerhead ribozyme sequence formed the 6-4-6-type helical

stem structure which was shown to provide optimal in vitro kinetic
values (8). Corresponding 33mer RNA hammerhead ribozymes
and 12mer RNA targets were chemically synthesized with 2-ACE
protection. (Dharmacon Research Inc, USA).
3.2. Ribozyme
Time-Course and
Multiturnover Kinetics
1. Target oligo RNAs are deprotected according to the

Dharmacons deprotection protocol.
Centrifuge tubes briefly. Add 400 mL of 2-deprotection buffer
to each tube of RNA. Add 800 mL of 2-deprotection buffer to
oligos with homopolymer stretches of rA longer than 12 bases.
Completely dissolve RNA pellet by pipetting up and down.
Vortex for 10 s and centrifuge for 10 s. Incubate at 60C for
30 min. Incubate at 60C for 2 h for oligos with biotin modifications or homopolymer stretches of rA longer than 12 bases.
Lyophilize or SpeedVac to dryness before use.
2. Label with g-[32P]-dATP using the following reaction: 2 mL
oligo-RNA (10 pmol/mL), 1 mL RNasin (Promega; Madison,
WI, USA), 1 mL 0.1 M DTT, 3 mL ddH2O, 1 mL [g32P]-dATP,
1 mL 10 PNK buffer, and 1 mL T4 polynucleotide kinase
(Roche Molecular Biochemicals; Indianapolis, IN, USA).
122
3. Incubate at 37C for 30 min, and then dilute to 100 mL with

ddH2O followed by extraction with phenol/chloroform/
isoamyl alcohol (25:24:1).
4. Free nucleotides are removed by passing the aqueous layer on
a Sephadex G25 fine spin column.
5. RNA is ethanol precipitated and resuspended in 100 mL ddH2O
to a final concentration of 0.2 pmol/mL.
6. Ribozyme cleavage reactions are performed in the presence or
absence of various concentrations of ribozyme and target RNA
in a reaction mix (20 mL) containing 40 mM Tris/HCl, pH
7.5, and 20 mM MgCl2. Samples are incubated at 37C and the
reaction is initiated by addition of ribozyme to target RNA.
7. At the appropriate times, the reactions are arrested with the
addition of a 6 mL of 90% formamide, 50 mM EDTA (pH
8.0), 0.05% xylene cyanol, and 0.05% bromophenol blue.
8. Reaction products are separated on a 1519% polyacrylamide
gel containing 8 M urea which can resolve RNA from 20 to
800 bases. The gel is run according to Invitrogen specifications
using 1 TBE running buffer. The gel is run at a constant voltage of 180 V for 5085 min. The samples are quantitated by
radioanalytic scanning (PhosphorImager; Molecular Dynamics,
Durham, NC, USA).
9. In the time-course study, reaction mixtures include 10 pmol
ribozyme and 100 pmol target RNA (containing 0.2 pmol
g-[32P]-target). Reactions are stopped at 0.5, 1, 2, 3, 4, 5, 10,
and 30 min, 1, 2, 3, and 15 h. In the multiturnover study, reactions are stopped at 1 min. Reactions include 0.015 pmol/mL of
ribozyme and increasing concentrations of target RNA (0.15
15 pmol/mL) as shown in Table 1 (see Note 1) (Fig. 2).
10. MichaelisMenton constant (Km) and reaction rate at saturating substrate concentration (kcat) are obtained using doublereciprocal plots of velocity versus substrate concentration.
The concentrations of target RNA range from 0.15 to
15 pmol/mL with a constant ribozyme RNA concentration of
0.015 pmol/mL (Fig. 3).
3.3. Plasmid
Construction
1. Ribozymes with the better kinetic properties for TGF-b1 or

CTGF are selected to test their efficiency in cells. The first step
is to synthesize a plasmid that expresses the ribozyme. Singlestranded synthetic DNA oligonucleotides encoding complementary sequences of the ribozyme are chemically synthesized.
A second pair of oligonucleotides is constructed which contains
a single-nucleotide replacement, shown as the underlined
nucleotides (C G, G C), to create an inactive ribozyme
that would assess the general toxicity and antisense effect of
the ribozyme.
0.15
Ribozyme
Target
0.15
0.015
Table 1
Multiturnover kinetics analysis
0.3
0.015
0.6
0.015
0.9
0.015
1.2
0.015
1.5
0.015
3.0
0.015
6.0
0.015
9.0
0.015
10
12
0.015
11
15
0.015
12
8
123
124
Fig. 2. Time-course studies of TGF-b ribozymes (Rz) cleaving target RNAs. Cleavage reactions were carried out at constant
concentrations of 10 pmol ribozyme and 100 pmol target, and were stopped and analyzed at 0.5, 1, 2, 3, 4, 5, 10, and
30 min, 1, 2, 3, and 15 h. Both of TGF-b1 Rz 1 and TGF-b1 Rz 2 could cut their targets. Prolonged incubations caused a
significant increase in cleaved product, although the reaction rates slowed markedly after 30 min. The TGF-b1 Rz 1 was
slightly more active than TGF-b1 Rz 2, cleaving 62% compared with 43% at 10 min, and 94% compared with 82% at the
end of incubation. Data was collected from three individual tests.
Fig. 3. Multiturnover studies of TGF-b1 Rz 1 and TGF-b1 Rz 2. The concentration of ribozymes and targets is indicated in
Table 1. The enzymatic reaction displayed reaction kinetics amenable to MichaelisMenten analysis. TGF-b1 Rz 1:
Km = 2.78 mM, kcat = 74.1/min. TGF-b1 Rz 2: Km = 12.50 mM, kcat = 92.2/min. Although TGF-b1 Rz 1 had lower Vmax and kcat
than that of TGF-b1 Rz 2, the lower kcat of this rbozyme is compensated by its lower Km value. The kcat/Km of TGF-b1 Rz 1
and TGF-b1 Rz 2 were separate, 2.7 l07/M/min and 7.4 106/M/min. Thus, TGF-b1 Rz 1 is 3.6 times more efficient.
125
2. The complementary oligonucleotides are annealed, producing

NsiI and HindIII restriction sites. The fragments are inserted
into the pTR-UF-21HP vector, which has been linearized with
HindIII and NsiI restriction enzymes. Synthesis of the
ribozyme is driven by the chicken b-actin promoter and CMV
enhancer in this vector. The presence and correct orientation
of the insert are verified by DNA sequencing. The pTRUF21HP vector contains a hairpin ribozyme following the
insert site that self-cleaves the mRNA when transcribed in
the cell, yielding a relatively short 3 arm of hammerhead
ribozyme that improves cleavage efficiency (7).
3.4. Analysis of
Endogenous Target
mRNA Knockdown
by a Ribozyme
3.4.1. CTGF Ribozyme
Analysis
Human Cell Culture
and Transfection
1. Cultures of human newborn foreskin fibroblasts (ATCC;

Manassas, VA, USA) are cultured in equal parts DMEM,
Medium 199, Hams F12 nutrient mixture containing 1 mM
NaHCO3, and buffered with 25 mM HEPES at pH 7.4. The
medium is supplemented with 10% heat-inactivated normal
calf serum and 1 antibioticantimycotic (Gibco BRL).
Exponentially growing cells are transfected with vector (pTRUF21), inactive ribozyme plasmid (pTR-UF21-In), or active
CTGF ribozyme plasmid (pTR-UF21-CHR745) using
Lipofecatime reagent (Invitrogen Life Technologies; Carlsbad,
CA, USA).
2. Since pTR-UF-21 contains a neomycin-resistance gene; cells
that are stably transfected are selected with 200 mg/mL geneticin (G418 Sulfate) added to the culture medium 48 h after
transfection. After 7 days, selected cells are transferred to
48-well plates for evaluation of ribozyme effects.
Quantitative Reverse
Transcription-Polymerase
Chain Reaction
1. Confluent cultures of stably transfected fibroblasts in 48-well

plates are held in serum-free medium for 48 h before RNA
extraction (Qiagen RNeasy Kit; Valencia, CA, USA). Cells are
stimulated for 24 h with 5 ng/mL human TGF-b1 to stimulate CTGF expression. Total RNA is extracted using TRIzol
reagent according to the manufacturers protocol.
2. CTGF mRNA transcripts are detected using the TaqMan realtime quantitative RT-PCR procedure (9). A standard curve is
generated using CTGF mRNA transcripts that are transcribed
in vitro using T7 RNA polymerase from a plasmid containing
CTGF cDNA. CTGF transcript is precipitated with ethanol
and dissolved in diethylpyrocarbonate (DEPC)-treated water.
3. Reactions are assembled in a 96-well optical reaction plate.
Each reaction contains 1 TaqMan One-step RT-PCR Master
Mix, 900 nM forward primer (5-AGCCGCCTCTGCAT
GGT-3), 900 nM reverse primer (5-CACTTCTTGCCCTTC
TTAATGGTTCT-3), 2 mM fluorescent TaqMan probe
( 5 - 6 F A M - T T C C A G G T C A G C T T C G C A A G G C C TTAMRA-3), and RNA sample (CTGF mRNA standard or
126
500 ng of sample RNA) to a final volume of 25 mL per reaction

(see Note 2). The plate is analyzed on the ABI Prism 5700
Sequence Detection System (Applied Biosystem, Foster City,
CA, USA), which simultaneously performs the RT-PCR and
detects a fluorescence signal. A standard curve is generated
using the transcribed CTGF mRNA samples (2.3 102 to
2.3 106 pmol). The level of GAPDH mRNA is also measured
in each sample using the TaqMan GAPDH Control Kit
(Applied Biosytems, Foster City, CA, USA), and the number
of CTGF mRNA molecules in samples is expressed as pmol
CTGF mRNA per pmol of GAPDH mRNA.
4. Levels of mRNA are expressed as mean standard error of six
replicate samples for each condition, and ANOVA and Tukeys
HSD post hoc test are used to assess statistical significance
between times and groups (Fig. 4).
CTGF Enzyme-Linked
Immunosorbent Assay
1. CTGF is measured in conditioned medium and in cytoplasmic

extracts of serum-starved, cultured cells following stimulation
for 24 h with 5 ng/mL human TGF-b1 using a capture sandwich enzyme-linked immunosorbent assay (ELISA) with biotinylated and nonbiotinylated, affinity-purified goat polyclonal
antibodies to human CTGF (10). A flat-bottom ELISA plate
(Costar 96-well) is coated with 50 mL of goat antihuman
CTGF antibody (which recognizes predominately epitopes in
the N-terminal half of the CTGF molecule) at a concentration
of 10 mg/mL in PBS/0.02% sodium azide for 1 h at 37C.
2. Wells are washed four times and incubated with 300 mL of
blocking buffer (PBS/0.02% sodium azide/1% bovine serum
albumin) for 1 h at room temperature (see Note 3).
3. The wells are washed four times and 50 mL of recombinant
human CTGF protein (from 0.1 to 100 ng/mL) or sample is
added and incubated at room temperature for 1 h.
4. After washing, 50 mL of biotinylated goat antihuman CTGF
(2 mg/mL) is added and incubated at room temperature in the
dark for 1 h, then washed, and 50 mL of alkaline phosphataseconjugated streptavidin is added and incubated at room
temperature for 1 h.
5. The wells are washed again and incubated with 100 mL of alkaline phosphatase substrate solution (1 mg/mL p-nitrophenyl
phosphate in sodium carbonate/bicarbonate buffer/0.02%
sodium azide, pH 9.6). Absorbance at 405 nm is measured
using a microplate reader (Molecular Devices, Sunnyvale, CA).
6. CTGF levels are normalized for total protein content of samples
using bicinchoninic acid (BCA) protein assay reagent (Pierce
Chemical, Rockford, IL, USA) and are expressed as ng/mg
127
Fig. 4. Effect of TGF ribozyme on expression in cell culture. (a) Effect of CTGF Rz 1 on CTGF mRNA expression in human
dermal fibroblast cultures. CTGF mRNA was then measured using TaqMan quantitative RT-PCR and results were normalized
to GAPDH mRNA. The level of CTGF mRNA expression in fibroblasts that were stably transfected with the plasmid expressing
CTGF Rz1 was decreased by 55% (p < 0.01, n = 6) compared with nontransfected control fibroblasts. In contrast, transfection of fibroblasts with the empty expression vector pTR-UF21 or with a plasmid expressing the catalytically inactive
ribozyme did not significantly alter the levels of CTGF mRNA from control cells. (b) Effect of CTGF Rz1 on CTGF protein
expression in human dermal fibroblast cultures. CTGF protein was measured in cytoplasmic extracts and conditioned
medium samples using CTGF sandwich ELISA and results were normalized for total protein concentration. The levels of
CTGF protein measured in conditioned medium of detergent extracts of fibroblasts expressing CTGF Rz1 were reduced by
72 and 71%, respectively, compared with nontransfected control fibroblasts control groups (p < 0.01, n = 6).
protein for six replicate samples for each condition. Sensitivity

of the ELISA is 0.1 ng/mL with an intra-assay variability of
3%, which is similar to a previously published ELISA for
CTGF (11).
7. Levels of protein are expressed as mean standard error of
six replicate samples for each condition, and ANOVA and
Tukeys HSD post hoc test are used to assess statistical significance between times and groups.
128
3.4.2. TGF-b1 Ribozyme

Analysis
Mouse-Immortalized Cell
Culture and Transfection
1. CCL-1 cells (NCTC clone 929, ATCC, USA) are cultured

with equal parts Hams F-12, Medium 199, and DMEM
media, containing 20 mM HEPES, 1 mM NaHCO3, 100 U/mL
penicillin, and 100 mg/mL streptomycin, supplemented with
10% normal goat serum at 37C.
2. Exponentially growing cells are trypsinized and resuspended in
hypoosmolar electroporation buffer (Eppendorf Scientific,
Inc., Germany). Cell density is adjusted to 106 cells/mL. Cell
suspensions are combined with pTR-UF-21-THRC576 or
pTR-UF-21-THR576 (10 mg/mL final concentration, in
ddH2O). 400 ml of the mixture is placed in electroporation
cuvettes (2-mm-gap width, Eppendorf). Following incubation
on ice for 10 min, the suspensions are electroporated in a
Multiporator (Eppendorf) at 1 pulse of 400 V for 50 ms.
Subsequently, the suspension is incubated in the cuvette for
510 min at room temperature. The suspension is then transferred from the cuvette into the culture medium and distributed into a 12-well plate.
3. Because pTR-UF-21 includes a gene for neomycin resistance,
cells that are stably transfected are selected using Geneticin
(G-418 Sulfate). 48 h after electroporation, G418 is added to
the culture medium at a concentration of 600 mg/mL, as
determined by a series of concentration tests. After 7 days, the
culture medium is changed and the G418 concentration is
decreased to 200 mg/mL to maintain selection. Cell monoclones
are trypsinized and transferred to another plate to continue
culturing in selecting culture medium (G418 200 mg/mL).
Nontransfected CCL-1 cells are used as a negative control.
Before examining RNA and protein expression, confluent cells
in 6-well plates are cultured with 2 mL serum-free medium
containing 200 mg/mL G418 for 48 h.
RNA Extraction and

Reverse TranscriptionPolymerase Chain Reaction
1. Total RNA is extracted using TRIzol reagent according to

the manufacturers protocol. Cells are washed with PBS three
times followed by addition of 1 mL TRIzol to each well to
lyse the cells.
2. Concentration and purity of RNA are measured using spectrophotometry at 260 nm (GeneQuant; Amersham Pharmacia
Biotech, Uppsala, Sweden). 260/280 nm ratios of all the samples should be 1.90 or greater.
3. Reverse transcription is performed using a first-strand synthesis
kit (Superscript TM First-strand synthesis system for RT-PCR,
GIBCO BRL) using oligodeoxythymidine primers. Relative
cDNA levels are quantitated by co-amplification of the
129
housekeeping gene beta-actin and TGFb1 using PCR Master

Mix (Promega). The primer sequences for beta-actin are as follows:
forward, 5-TGCGTGACATTAAGGAGAAG-3 and reverse
5-GAAGGTAGTTTCGTGGATGC-3. The primer sequences
for TGFb1 are forward primer 5-GAAGCGCATCGAAGC
ATCC-3 and reverse primer 5-TTGGACAACTGCTCCA
CCTT-3.
4. Since oligo-dT is used to perform the reverse transcription, the
PCR products indicated intact RNA levels. Amplification is
performed in a 50 mL reaction mixture using the following
conditions: 1 mL of each primer (10 pmol/mL), 2 mL RT reaction
product samples, 25 mL PCR master mix, and 19 mL ddH2O.
PCR amplifications are initiated at 94C for 2 min followed by
28 sequential cycles of denaturation at 94C for 30 s, annealing
at 56C for 1 min, and extension at 72C for 1.5 min. A final
extension cycle at 72C for 10 min is performed in a thermocycler (Twin blockTH System, San Diego, CA).
5. A video imaging and densitometry software system (Kodak
digital science, Eastman Kodak Company) is used to quantify the relative band intensities of beta-actin and TGFb1.
TGFb1 mRNA levels are then expressed relative to betaactin (Fig. 5).
TGFb-1 Enzyme-Linked
Immunosorbent Assay
The protocol for the TGF-b1 ELISA is same as that previously

described from the CTGF, except the antibodies used are specific
for TGF-b1.
3.5. Analysis
of Exogenous
Synthetic Target
Knockdown by a
Ribozyme (TGF-b1)
1. A secreted alkaline phosphatase (sAP) reporter gene driven by

an hEF1HTLV promoter is cloned into pBluescript and
contains a multiple cloning site located upstream of the sAP
reporter gene.
3.5.1. Production
of Secreted Alkaline
Phosphatase Target
Expression Plasmid
3.5.2. Human Embryonic
Kidney298 Cell Culture
and Dual Transfection
2. A target sequence, approximately 300 bp, containing the

ribozyme target sequence is cloned into the multiple cloning
site upstream of the sAP reporter gene.
1. DMEM, with 4.5 g/L glucose and 1 g/L L-glutamine:

Medium is supplemented with 10% heat-inactivated normal
calf serum and 1 antibioticantimycotic (Gibco BRL).
2. Exponentially growing cells are transfected with two plasmids,
the first being the psAP Bluescript TGF-b1 target plasmid and
the second being a plasmid that expresses one of the following:
green fluorescent protein (GFP), inactive TGF-b1 ribozyme,
130
Fig. 5. (a) The effect of TGF-b1 Rz 1 reducing TGF-b1 expression was tested by intracontrol RT-PCR. Bands of 213 bps were
the housekeeping gene beta-actin and bands at 1,014 bps were TGF-b1 expression in cells. Results showed that TGF-b1
expression in TGF-b1 Rz 1 transfected cells was significantly depressed by comparing with negative and inactive control
groups; the decreasing rates are 16.2 and 12.1%, respectively. Semiquantitative study shows that in TGF-b1 Rz 1 transfected cells the TGF-b1 expression was significantly depressed compared with the control groups (p < 0.01, n = 4). (b)
Testing the efficiency of TGF-b1 Rz 1 depressed the TGF-b1 protein expression in cytoplasm and culture medium supernatant. Protein expression was reduced by 59 and 37% in cytoplasm and conditioned medium, respectively. Compared
with control groups, TGF-b1 Rz 1 depressed the TGF-b1 expression significantly, both in cytoplasm and culture medium
supernatant (p < 0.01, n = 4).
and active TGF-b1 ribozyme. Turbofect reagent (Fermentas

Inc.; Glen Burnie, MD, USA) is used for the dual transfection
following the manufacturers protocol (see Note 4).
3. Forty-eight hours after transfection, concentration of sAP protein is assessed using Quanti-BlueTM (InvivoGen, San Diego,
CA, USA). Ribozyme activity levels are expressed as relative
sAP of GFP expression vector (Fig. 6).
131
Fig. 6. The effect of TGF-b1 Rz 1 on exogenous synthetic target as reported by a secreted

alkaline phosphatase gene. HEK293 cells were simultaneously transfected with two plasmids,
one expressing the TGF-b1 ribozyme target sequence with secreted alkaline phosphatase
reporter and the other a plasmid that expresses one of the following: GFP, inactive TGF-b1
ribozyme, or active TGF-b1 ribozyme. A statistically significant difference of 27% was
found when comparing the GFP plasmid with the active TGF-b1 ribozyme plasmid.
4. Notes
1. When setting up kinetic experiments, be sure to make reaction
mix with the substrate, but wait to add the ribozyme. Also,
bring the reaction mix containing the substrate to 37C and
then add the ribozyme. If you do not do this, your reaction
will need to be heated up to 37C and the catalytic activity will
be reduced.
2. When performing PCR, always mix the reagents by lightly
flicking and quickly centrifuge to bring the reagents to the bottom of the tube.
3. When performing ELISA, cover the 96-well plate during the
incubations. Also, when washing, gently tap the 96-well plate
on a paper towel to remove all of the wash solution.
4. From our experience, when doing a dual transfection using
Turbofect, a ratio of 1:1 of plasmids gave the greatest expression of both plasmids.
132
References
1. Border, W.A., Noble, N.A., Yamamoto, T.,
Harper, J.R., Yamaguchi, Y., Pierschbacher,
M.D. Ruoslahti, E. (1992) Natural inhibitor of
transforming growth factor-beta protects
against scarring in experimental kidney disease.
Nature 360: 361364.
2. Border, W. A., Noble, N. A. (1994) Transforming Growth Factor B in Tissue Fibrosis.
N. Engl. J. Med. 331: 12861292.
3. Grotendorst, G.R., Duncan, M.R. (2005)
Individual domains of connective tissue growth
factor regulate fibroblast proliferation and myofibroblast differentiation. FASEB J. 19: 729738.
4. Lewin, A., Hauswirth, W. (2007) Ribozyme
gene therapy: applications for molecular medicine. Trends Mol. Med. 7: 221228.
5. Lee, P.A., Blatt, L.M., Blanchard, K.S.,
Bouhana, K.S., Pavco, P.A., Bellon, L.,
Sandberg, J.A. (2000) Pharmacokinetics and
tissue distribution of a ribozyme directed
against hepatitis C virus RNA following subcutaneous or intravenous administration in mice.
Hepatology 32: 640646.
6. Shimayama, T., Nishikawa, S., Taira, K. (1995)
Generality of the NUX rule: kinetic analysis of
the results of systematic mutations in the trinucleotide at the cleavage site of hammerhead
ribozymes. Biochemistry 34: 36493654.
7. Fritz, J.J., Lewin, A., Hauswirth, W., Agarwal,

A., Grant, M., Shaw, L. (2002) Development
of hammerhead ribozymes to modulate endogenous gene expression for functional studies.
Methods 28: 276285.
8. Drenser, K.A., Timmers, A.M., Hauswirth,
W.W., Lewin, A.S. (1998) Ribozyme-targeted
destruction of RNA associated with autosomaldominant
retinitis
pigmentosa.
Invest.
Ophthalmol. Vis. Sci. 39: 681689.
9. Heid, C.A., Stevens, J., Livak, K.J., Williams,
D.M. (1996) Real time quantitative PCR.
Genome Res. 10: 984986.
10. Blalock, T.D., Duncan, M.R., Varela, J.C.,
Goldstein, M.H., Tuli, S.S., Grotendorst, G.R.,
Schultz, G.S. (2003) Connective tissue growth
factor expression and action in human corneal
fibroblast cultures and rat corneas after photorefractive keratectomy. Invest. Ophthalmol. Vis.
Sci. 44: 18791887.
11. Tamatani, T., Kobayashi, H., Tezuka, K.,
Sakamoto, S., Suzuki, K., Nakanishi, T.,
Takigawa, M., Miyano, T. (1998) Establishment
of the enzyme-linked immunosorbent assay
for connective tissue growth factor (CTGF)
and its detection in the sera of biliary atresia.
Biochem. Biophys. Res. Commun. 251:
748752.
Chapter 9
Control of the Interferon Response in RNAi Experiments
Jana Nejepinska, Matyas Flemr, and Petr Svoboda
Abstract
The RNA interference (RNAi) and interferons have been an uneasy marriage. Ever since the discovery of
RNAi in mammals, the interferon response has been a feared problem. While RNAi became an efficient
and widespread method for gene silencing in mammals, numerous studies recognized several obstacles,
including undesirable activation of the interferon response, which need to be overcome to achieve a specific
and robust RNAi effect. The aim of this text is to provide theoretical and practical information for scientists
who want to control interferon response and other adverse effects in their RNAi experiments.
Key words: RNA interference, Small interfering RNA, Short hairpin RNA, Double-stranded RNA,
Interferon
1. Introduction
RNAi is an excellent tool for selective inhibition of gene expression
and studies of gene function(s) when properly used. Otherwise,
it is an excellent tool to generate confusing results. While RNAi
became a standard tool, the lack of appropriate controls and/or
ignorance of nonspecific effects undermined its efficient use in various cases. This text gives a brief overview of adverse effects found
in RNAi experiments in mammalian cells and provides guidelines
for designing RNAi experiments and identifying one of the frequently
encountered undesirable effects the interferon response.
1.1. RNA Silencing
in Mammals and Its
Experimental Use
RNA interference (RNAi) and the microRNA (miRNA) pathways

regulate gene expression by inducing sequence-specific degradation and/or translational repression of target mRNAs (reviewed
for example in refs. 14). A common feature of both pathways is
2122 nucleotide-long RNA molecules serving as sequence-specific
133
134
J. Nejepinska et al.
nucleus
miRNA pathway
cytoplasm
RNAi
pathway
Exportin 5 mediated transport

Drosha
dsRNA
pre-miRNA
pri-miRNA
Dcr
Class 2 hairpin
(miRNA-like)
HAIRPIN
EXPRESSION
VECTORS
.
TRANSFECTION
short RNAs
(miRNAs and siRNAs)
Class1 hairpin
(shRNA)
siRNA
RISC
loading
Ago
miRNA pathway
RELOCATION TO
P-BODIES
RNAipathway
Ago
mRNA degradation
Ago2
AAAAA
Inhibition of translation
AAAAA
Cleavage of
mRNA by Ago2
Fig. 1. RNA silencing in mammalian cells and its experimental use. The miRNA and RNAi pathways and two mechanisms
of post-transcriptional silencing are depicted. Note that following Dicer cleavage, the miRNAs and siRNAs share a common
pathway. Thus, the final silencing effect is dependent on the degree of homology with a cognate mRNA and the nature of
an AGO protein rather than on the origin of the short RNA. The common entry points for experimental activation of RNAi are
indicated in blue.
guides for silencing (Fig. 1). These short RNAs are released by
Dicer, an RNase III family endonuclease, from various forms of
double-stranded RNA (dsRNA). Mammals have only one Dicer
protein, common for both RNAi and miRNA pathways.
The classical RNAi is initiated by long perfect dsRNA, which is
processed by Dicer into double-stranded short interfering RNAs
(siRNAs). siRNAs perfectly base-pair with a cognate RNA and
guide its cleavage in the middle of the base-pairing sequence.
Delivery of chemically synthesized siRNAs into mammalian cells
also induces sequence-specific knockdown (5). However, long
dsRNA is likely not a natural substrate of Dicer in mammalian somatic
cells as dsRNA >30 bp is known to trigger sequence-independent
pathways such as the protein kinase R (PKR) pathway (6). Long
dsRNA induces RNAi only in oocytes, early embryos, embryonic
stem cells, and possibly a few other mammalian cell types (7). Small
RNA cloning experiments discovered that virtually all endogenous
short RNAs linked to RNA silencing in somatic mammalian cells
135
are miRNAs (e.g., refs. 8, 9). miRNAs are transcribed as long

primary transcripts (pre-miRNAs) with local hairpin structures that
are processed by a nuclear RNase III Drosha-containing complex
into short hairpin intermediates (pre-miRNAs). Pre-miRNAs are
transported to the cytoplasm where Dicer releases a duplex containing
a miRNA. A typical mammalian miRNA imperfectly base-pairs with
a cognate 3UTR and inhibits protein translation.
Despite certain distinctions, mammalian RNAi and miRNA
pathways can be seen as one biochemical RNA silencing pathway
because the effector complexes loaded by Dicer products appear
functionally similar if not identical. Both siRNAs and miRNAs are
loaded onto an Argonaute-containing effector ribonucleoprotein
(RNP) complex, referred to as miRNP or RISC (RNA-induced
Silencing Complex), which executes silencing. Four mammalian AGO
proteins (AGO1 through AGO4) associate with miRNAs and are
implicated in translational repression (1012). In addition,
AGO2 can mediate endonucleolytic cleavage of a target mRNA in
the middle of the base-paired sequence (10, 11, 13). The AGO2mediated cleavage requires formation of a perfect RNA duplex,
while imperfect base-pairing, typical for most miRNAs, generally
results in translational repression (14, 15). However, examples of
miRNAs inducing RNAi-like cleavage also exist (16). Therefore,
whether a short RNA will cause RNAi-like endonucleolytic cleavage
or will induce the translational repression as a miRNA depends on
the degree of complementarity and the AGO protein present,
rather than on the origin of the short RNA. Despite this overlap
between miRNA pathway and RNAi in mammals, we use the term
RNAi for any experimental induction of sequence-specific cleavage,
even if triggered by miRNA-like RNAs.
Structurally, there are three categories of short RNAs inducing
RNAi, commonly referred here as RNAi triggers (Fig. 2): (1) siRNA
Duplexes of 21-22-mers with two nucleotide 3 overhangs.
(2) Class I short hairpin Based on covalent linking of strands carrying
functional siRNA sequences. The minimal class I hairpin contains a
19-bp dsRNA stem and 49 nucleotide loop and it is probably not
processed like a classical miRNA (1720). (3) Class II hairpin
Directly modeled after pre-miRNA (18, 21, 22).
RNA triggers can be prepared in vitro or expressed from DNA.
The two most common experimental designs are (1) transient
transfection of commercially obtained siRNAs and (2) transient or
stable transfection of vectors expressing class I short hairpins from
a pol III promoter. Class II hairpins were used less frequently but
they have recently come into focus because they can be expressed
together with a reporter from a single pol II promoter, thus providing more versatility than pol III-driven systems. It is important
to mention these strategies because some of the nonspecific effects
in RNAi experiments can be attributed to the carrying vector or
delivery method.
136
siRNA
class I shRNA
class II shRNA
cationic lipid-RNA complexes

activate IFN via TLR3 and TLR7
5 triphosphate
introduced by phage
RNA polymerases
activates IFN
siRNA < 30 bp can activate PKR

dsRNA > 30 bp activates PKR and 2,5-OAS
some sequence
motifs within ssRNA
can activate IFN
lack of 3 overhangs
induces IFN via Rig-I
Fig. 2. Structural features of RNAi triggers and their relationship to interferon activation.
(a) Three different types of RNAi triggers targeting firefly luciferase. siRNA and class II
shRNAs were found in the literature (5, 56). Class I hairpin was modeled according to
Brummelkamp et al. (17). Note that class I shRNA sequence represents the whole transcript
produced by pol III from a vector while the class II shRNA corresponds to the part of the
pol II transcript processed by Drosha and Dicer. Sequence within the gray box indicates
the RNA fragment incorporated into the RISC complex. (b) Schematic display of structural
features of an siRNA, which can trigger interferon response. See text for original references.
1.2. Nonspecific
Effects in RNAi
Experiments
Nonspecific effects can be divided into three categories. The first

category includes the effects caused by activating interferon-related
pathways. These effects are usually independent of siRNA sequence
and they involve transcriptional activation of interferon-stimulated
genes (ISGs). Although we generally refer to these effects as an
interferon response, it should be kept in mind that an interferon
response is broader and includes other pathways and effects not
discussed here. The second category, referred to as off-targeting
(see Subheading 1.4), includes sequence-dependent effects caused
by targeting unintended transcripts. Finally, excessive use of RNAi
137
triggers may cause unintended effects by saturating the endogenous

miRNA pathway resulting in relief of repression of genes repressed
by miRNAs.
1.3. Interferon
Response
Mammals have a complex system for responding to dsRNA in the

cytoplasm. Various forms of cytosolic dsRNA can interact with
several proteins that take part in the innate immune response. The
interferon pathway is the most ubiquitous sequence-independent
pathway induced by dsRNA in mammalian cells (reviewed in detail
for example in ref. 23). One of the best-characterized effects of
dsRNA is activation of PKR, which phosphorylates translation
initiation factor eIF2a and causes general repression of translation.
PKR is also involved in the regulation of NF-kB, which plays a
key role in interferon induction. Interferon and dsRNA also activate 2,5-oligoadenylate synthetase (2,5-OAS) that produces
2,5-oligoadenylates with 5-terminal triphosphate residues that
subsequently induce activation of RNase L, a protein responsible
for general RNA degradation (23).
Experimental induction of RNAi may result in activation of
interferon response through some of the aforementioned proteins
but there are also mechanisms activating interferons in dsRNAindependent manner. Different stimuli can lead to activation of
overlapping but distinct sets of ISGs (24), and in specific cell types,
particularly immune cells, the interferon response can be elicited
by additional pathways (reviewed in ref. 25).
There are diverse features of RNAi triggers that can lead to the
interferon activation through some of the proteins recognizing
dsRNA. In 2003, two groups reported that transfection of siRNAs
as well as pol III-driven shRNA expression can activate the interferon response (26, 27). Sledz et al. reported that transfection of
siRNAs using Oligofectamine into different mammalian cells
induced an RNAi effect as well as activation of PKR and expression
of numerous ISGs (26). Microarray profiling of cells transfected
with different concentrations (10, 25, 50 and 100 nM) of siRNAs
revealed approximately 50 ISGs induced more than twofold at 48 h
post-transfection. Some ISGs were induced at all siRNA concentrations
while others only at higher ones (26). However, even mock transfection alone can sometimes have a stimulatory effect on ISGs (28).
A detailed analysis of these effects revealed that siRNAs lacking
2-nt 3 overhangs activate the interferon system via RNA helicase
RIG-I (29). Thus, transfection of higher siRNA amounts may
result in appearance of interferon-stimulating small RNAs lacking
2-nt 3 overhangs, which seem to be a structural basis for discriminating between Dicer products and other short dsRNAs.
Another mechanism of activating interferons in RNAi experiments
was described for short hairpins or siRNAs generated by in vitro
transcription with phage polymerases (30). In this case, the interferon
response is induced by the 5 triphosphate GTP created by the
138
phage polymerase. Interferon induction by the triphosphate RNA

ends is also mediated by RIG-I (31, 32). Therefore, whenever T7
or other phage polymerases are used to produce siRNAs or shRNAs,
it is advisable to appropriately process their 5 termini and include
controls for measuring the interferon induction.
The interferon response was also induced by type I shRNAs
expressed from H1 or U6 promoters (27, 33). Analysis of
U6-promoter vectors, which induced strong ISG activation, identified
a critical AA dinucleotide motif near the transcription start site suggesting a design flaw in U6-promoter based vectors (33). The exact
mechanism of how the presence of the AA motif induced ISGs is
unknown but these results provide a rationale for a better design of
U6-driven shRNA vectors. Plasmids with H1 promoter too is not
immune to ISG induction, but its exact cause remains unknown
(33). Notably, interferon activation may not be caused only by
shRNA expression only as lentiviral vectors seem to be less likely to
induce OAS (one of the ISGs) than plasmid vectors (27).
Finally, interferon induction can be siRNA-specific as several
motifs within single-stranded RNAs (ssRNA) from siRNAs, such as
UGUGU and GUCCUUCAA, stimulate the interferon response in
immune cells (34, 35). In addition, there are also immunostimulatory siRNAs without defined sequence motifs (36). Activation of
the interferon response in these cases was likely mediated by TLR
receptors and linked to endosomes (reviewed in detail in ref. 37).
Most researchers aim to avoid interferon activation while
achieving an RNAi effect. Their interest in mechanisms of interferon
activation ends when the interferon response is absent in their
experiments. One of the common approaches to detect induction
of the interferon pathways in cell lysates or even culture media is
to employ antibodies recognizing ISGs (see Subheading 3.5.2).
However, this approach is less sensitive than the analysis of transcripts
of the interferon pathway and it may become costly. Therefore,
RT-PCR analysis of some of the ISGs is the most accessible approach
to detect if the interferon response was elicited in an experiment
(see Subheading 3.5.1). Finally, if one is using microarrays to
analyze results of an RNAi experiment, examination of probes
detecting ISGs will provide a good indication whether or not the
interferon response occurred (see Subheading 3.5.3).
1.4. Off-Targeting
Off-targeting occurs because short RNAs intended to induce

Ago2-mediated cleavage of perfectly base-pairing targets imperfectly hybridize to other transcripts. One can view a short RNA
introduced into a cell as a novel, abundant miRNA for which the
cellular transcriptome is not adapted. This implies that off-targeting
is common and can be found in most if not all RNAi experiments.
Even worse, off-targeting cannot be effectively predicted as the
reliability of computational miRNA target prediction is poor and
139
heavily dependent on conservation of miRNA binding sites. But,

as we discuss later, off-targeting can be reduced by certain modifications of siRNAs and experimental design and its potentially
destructive effect can be reduced by appropriate controls and
careful interpretation of results.
The extent of off-targeting was not recognized in initial RNAi
experiments. The early studies suggested that RNAi silencing
requires perfect base-pairing. A sufficient control for RNAi specificity
seemed to be an unrelated gene, typically a well-expressed housekeeping gene. Such control would indicate a global repression of
gene expression caused for example by global repression of translation and/or nonspecific mRNA degradation: the hallmarks of PKR
and 2,5-OAS activation. However, such controls are unlikely to
detect off-targeting. Even if off-targeting would affect expression
levels of hundreds of genes in a cell, the chance that one selected
marker gene would be affected is slim.
The first strong evidence of off-targeting effects was demonstrated when mammalian cells transfected with siRNAs were
systematically analyzed using microarrays (38). Jackson et al. found
siRNA-specific expression patterns in transfected cells with only a
few genes regulated in common by different siRNAs against the
same gene. Although the effect was decreased when siRNA concentrations were lowered, the off-target regulation could not be
eliminated completely and many of the off-targeted genes showed
similar kinetics of targeting as the intended target. A similar effect
has been observed with an siRNA targeting a luciferase sequence
that has no homology in human genome. Moreover, off-target
effects directed by the passive siRNA strand have also been detected.
Although off-target regulation could not be completely explained,
a portion of it appeared to be caused by partial complementarity
between an siRNA and its target, reminiscent of the 5 seeding
regions of miRNAs (39). Off-targeting with various forms of
RNAi triggers has been repeatedly demonstrated (38, 40) and
some degree of off-targeting is likely widespread in RNAi experiments. However, improved understanding of siRNA and miRNA
target recognition, better siRNA chemistry and pooling provide
measures allowing for elimination of off-target effects in future
experiments (41).
1.5. Controlling
Nonspecific Effects
in RNAi Experiments
The nonspecific effects in RNAi experiments are common and the

best way to deal with them is to proper experimental design
(42, 43). It should become a common policy to accept only those
RNAi experiments that include controls truly decreasing the probability of misinterpretation due to nonspecific effects. An ideal
RNAi experiment should (1) include one or more sensitive markers
for the interferon response, (2) use two or more different siRNAs
140
targeting the same gene, (3) contain a rescue control by expressing

an RNAi resistant version of the targeted gene, and (4) use phenotypic
analysis designed to yield as uncommon phenotype(s) as possible.
2. Materials
2.1. Cell Lines
and Culture Media
Selection of cell lines depends on individual needs or preferences.

Common cell lines can be obtained from the American Type
Culture Collection or other commercial sources. The following
protocols are based on experience with the commonly used HeLa
and HEK293 cell lines. These cells were maintained in Dulbeccos
modified Eagles medium (DMEM, Invitrogen) containing 10%
fetal calf serum (FCS, Invitrogen), penicillin (100 U/mL, Invitrogen),
and streptomycin (100 mg/mL, Invitrogen).
2.2. siRNAs
There are a number of commercial sources offering siRNA synthesis.

They also provide predesigned, and in some cases even verified
siRNAs. Among the most common providers are Thermo Fisher
Scientific (former Dharmacon), Ambion, Qiagen, and Sigma. We
recommend searching Web sites of these providers for information
concerning predesigned and validated siRNAs targeting gene(s) of
interest. One of the attractive options for RNAi knockdown is to
use pools of siRNAs. These can be prepared in vitro from long
dsRNA substrates (esiRNA) using own protocol or some of the
commercial kits. Or, one can purchase a pool directly from a vendor.
Particularly attractive option is to use ON-TARGET plus SMART
pool siRNAs (Thermo Fisher Scientific), which reduce off-targeting
by pooling siRNAs (see Note 3), which are in addition modified at
their 5end to reduce miRNA-like behavior (44).
2.3. DNA
Oligonucleotides
These should be obtained from a local provider. As mutations in

in vitro synthesized oligonucleotides are a common problem
during cloning, we highly recommend using purified oligonucleotides from a provider with a good record of producing long DNA
oligonucleotides.
2.4. shRNA Expressing

Vectors
There are a number of different commercial and noncommercial

plasmid vectors for RNAi that are accessible to individual researchers,
and it is certainly a worthwhile investment to test several different
vectors before committing resources to a specific one. The choice
of the vector depends on whether one plans to make transient or
stable transfections or to have an inducible or tissue-specific knockdown. Thus, refer to the recent literature and information at manufacturers Web sites to choose a suitable experimental setup. The
following protocols are designed for inserting oligonucleotides
to produce Type I small hairpin from pSuper (OligoEngine) and
141
its derivates, such as pTer (20) (Bgl II/Hind III cloning sites) or
Type II miRNA-like small hairpins from pTMP or pLMP plasmids
(Open Biosystems) (EcoRI/XhoI cloning sites). However, we want
to point out that, while we and our collaborators routinely use
these vectors, other vectors are not necessarily inferior.
To verify the sequence of inserted oligonucleotides, pSuper
derivates can be sequenced with T3, T7 and M13 primers. Refer
to the exact map of a pSuper derivate to select a suitable primer.
For sequencing inserts in pTMP/pLMP vectors, we use the following
primers, which are localized upstream of the hairpin insertion
site: pTMP: 5-TTGACCTCCATAGAAGACACCG-3, pLMP:
5-CCTCATCACCCAGGTTAAGAT-3.
2.5. Reagents
1. Restriction enzymes (Fermentas, New England Biolabs, 10 U/mL)

with buffers: BglII and HindIII for pSuper derivates or XhoI
and EcoRI for pTMP and pLMP.
2. Agarose (e.g., Invitrogen), and electrophoresis running buffer.
We use SB buffer: 10 mM NaOH, 36 mM boric acid.
3. T4 DNA ligase with 10 ligation buffer (Fermentas).
4. Chemically competent Escherichia coli cells (e.g., DH5a strain).
shRNA-expressing vectors are usually readily propagated in
normal lab strains. However, inverted repeats in plasmids occasionally cause complications. In such a case, one can use strains,
which are able to maintain DNA with potentially highly structured
sequences, such as Sure (Stratagene) or Stbl4 cells (Invitrogen).
5. LuriaBertani (LB) medium [possibly terrific broth (TB) medium
for more yield).
6. LB agar plates: 1.5% Agar in LB medium with 100 mg/mL
ampicillin.
7. TE buffer: 10 mM TrisHCl (pH 7.5), 1 mM EDTA.
8. Ampicillin: Stock solution 100 mg/mL in water, working
concentration 100 mg/mL.
9. 60% Glycerol, sterile.
10. Gel extraction kit (e.g., QIAquick Gel Extraction Kit).
11. Miniprep kit (e.g., QIAprep Spin Miniprep kit).
12. MIDI or MAXIprep kit (e.g., Qiagen HiSpeed Plasmid Midi
or Maxi kit).
13. Transfection reagents for plasmid transfection: Turbofect
(Fermentas).
14. Transfection reagent for siRNA: Turbofect (Fermentas) or
Oligofectamine + OptiMEM (Invitrogen).
15. 30% Fetal calf serum (FCS): A stock made of 50 mL FCS,
1.7 mL 200 mM glutamine, 1.7 mL penicilin (10,000 U/mL)/
streptomycin (10 mg/mL), and 113.6 mL DMEM.
142
16. PolyI:C (Sigma), 5 mg/mL stock in water.

17. Plasmid expressing immunostimulatory shRNA, 200 ng/mL.
3. Methods
The following protocol is a basic protocol we use for RNAi experiments. Analysis of the interferon response is a compilation of data found
in the literature, our experience with occasional appearance of
interferon response in RNAi experiments (28), and ongoing analysis
of effects of long dsRNA expression in mammalian somatic cells.
3.1. siRNA Sequence
Design
Time consideration: 2 h
Proper siRNA design is crucial for conducting a successful RNAi
experiment. Incorrect siRNA sequence will result in decreased
siRNA efficiency and/or specificity of the silencing effect. Own siRNA
design carries a risk of a failure and it is not unusual when only
one of three, or even four siRNAs induces good repression of the
cognate gene. Ideally, one should obtain two siRNAs of different
sequences targeting the same gene, which sometimes makes siRNA
design a painstaking process. If only a single siRNA is available, one
of the controls should include a rescue experiment (see also
Subheading 1.5 and Note 2). Therefore, prior to making own
siRNA design, it is very useful to search PubMed and WWW (particularly siRNA vendor sites) to find whether suitable functional siRNA
sequences are available already.
Several important criteria for siRNA design have been identified
and they were built into a number of freely available Web-based
design tools (summarized for example in ref. 45). Preference of
one siRNA design tool over another is to some extent a matter
of personal choice (reviewed in ref. 46). We combine two design
tools: BIOPREDsi (47) and RNAxs (48) and we subsequently
verify the specificity of siRNAs using the Specificity Server (49).
BIOPREDsi is a neural network-based algorithm that has been
trained on a large set of siRNAs and has been used for a genomewide design of siRNA. BIOPREDsi was a top-scoring approach in
a comparative study of several common siRNA design tools (50)
and it is routinely used by us or our collaborators. BIOPREDsi is a
representative of tools that design siRNAs according to optimized
parameters of an siRNA sequence but not taking into an account
the interaction of an siRNA with its cognate mRNA. Therefore, we
complement the BIOPREDsi siRNA prediction with RNAxs tool,
which introduces the analysis of the cognate sequence accessibility.
RNAxs predicts secondary structures within the siRNA binding
site and evaluates probability of efficient recognition of the binding
site by the RISC complex, as it has been shown in biochemical
studies (48, 51).
143
1. Obtain the target mRNA sequence from NCBI (http://www.

ncbi.nlm.nih.gov) or ENSEMBL (http://www.ensembl.org).
For a simple knockdown, it is often recommended to use the
coding region (CDS) of target mRNA to design siRNAs
because the unique coding sequence reduces off-target risk.
The downside of using CDS for siRNA design is that a rescue
expression construct must carry specifically positioned mutations in the cognate sequence (see the section controls). When
targeting a 3 UTR, one can just use a different 3 UTR to
make a rescue expression plasmid.
2. Search for suitable siRNAs at the BIOPREDsi page (http://www.
BIOPREDsi.org). Paste the target mRNA sequence in FASTA
format (use Readseq to convert the sequence into FASTA format.
Readseq online can be found at different sites, we typically
use the one available in Sequence utilities at the BCM Search
Launcher site (http://searchlauncher.bcm.tmc.edu/)). Make
sure to have the Input Type set to RNA Sequence. Set # of
predicted siRNAs to 20, click Design siRNA sequences and
save the results. (Note: the siRNA design using BIOPREDsi
might be a little more time-consuming). If the Web site is not
available, you can use another similar siRNA designer Web site,
e.g., http://www.dharmacon.com/designcenter/.
3. Repeat the same procedure with the RNAxs (http://rna.tbi.univie.ac.at/cgi-bin/RNAxs). Set Maximal number of siRNAs
in the Output option to 20 again and click REPRESS IT.
4. Pick the sequences ranked from best to worst in RNAxs that also
have high scores from the BIOPREDsi prediction. Ideally, at
least four siRNAs with high scores should be selected. Verify that
these sequences are specific using the Specificity Server (http://
informatics-eskitis.griffith.edu.au/SpecificityServer (49)). Paste
the RefSeq code (NM_XXXXXX) of the target mRNA and the
siRNA sequence to the Option B Database search for matches
window and press Search. The server returns Automatic recommendation OK when the inspected siRNA passes specificity
criteria. Similarly, test the specificity of all selected sequences.
5. Optional: To test if siRNAs will be designed with high scores
in other siRNA design tools, one can use, for example DSIR
(http://cbio.ensmp.fr/dsir/) (52). An extensive list of other
siRNA design tools can be found elsewhere (45).
3.2. Production
of Vectors Expressing
Desired shRNAs
3.2.1. pTer/pSuper shRNA
Vector Cloning
Time consideration: 12 weeks
1. Design and obtain the sense and antisense oligonucleotides as

described in Fig. 3a.
2. Oligonucleotide annealing: Mix 5 mg of sense oligonucleotide
with 5 mg of antisense oligonucleotide in TE buffer in a total
volume of 100 mL. Place the tube with oligonucleotide mixture
144
Designed 19 nt siRNA sense strand
(identical with target mRNA sequence)
Designed 19 nt siRNA antisense strand

(complementary to target mRNA sequence)
Sense oligo
Antisense oligo
BglII
H1 promoter
HindIII
BGH pA
TRE
SV40 ori
pTER
4500bp
Zeocin
Ampicillin
Designed 19 nt siRNA sense strand

(identical with target mRNA sequence)
Designed 19 nt siRNA antisense strand

(complementary to target mRNA sequence)
Sense oligo
Antisense oligo
XhoI
EcoRI
5 miR30
min CMV
3 miR30
pgk
E
TR
5LTR and
Puromycin
TMP
8238bp
IRES
Ampicillin
GFP
3LTR
ORI
Fig. 3. Schematic overview of pTer (a) and TMP (b) vectors. Restriction sites for insertion of oligonucleotides carrying shRNA
sequences are visualized. The oligonucleotides should carry sense and antisense strands of the in silico designed 19-nt
siRNA as indicated in (a) for pTer and in (b) for TMP.
145
in a beaker with a large volume of boiling water (5001,000 mL)

and incubate for 2 min. Turn heating off and leave the oligonucleotide mix in hot water to cool down slowly. Dilute annealed
oligonucleotide 100 with TE buffer.
3. pTer vector digestion: Mix 2 mg of pTer plasmid DNA with
2 mL of 10 restriction buffer and add water to a final volume
of 18 mL. Add 1 mL of BglII enzyme and 1 mL HindIII enzyme
and incubate for 2 h at 37C.
4. Resolve the digested plasmid in a 1.0% agarose gel in SB buffer
and extract DNA from the gel using a commercial gel extraction kit according to manufacturers recommendations (e.g.,
QIAquick Gel Extraction Kit). Elute the extracted DNA in
30 mL of elution buffer.
5. Ligation: Mix 2 mL of digested pTer DNA with 5 mL of annealed
diluted oligonucleotide (1 ng/mL); add 2 mL of 10 ligation
buffer and nuclease-free water to a final volume of 19 mL. Add
1 mL of T4 DNA ligase and incubate for 2 h at room temperature
(RT) or overnight at 16C.
6. Transformation: Add 2 mL of the ligation mixture to 50 mL
of chemically competent E. coli cells (e.g., DH5a strain) and
incubate for 20 min on ice. Submit cells to a 30 s heat-shock at
42C and immediately add 1 mL of LB medium. Incubate for
1 h at 37C with shaking.
7. Selection: Plate 100 mL of transformed cells on LB agar
plate supplemented with 100 mg/mL ampicillin and incubate
overnight at 37C.
8. Vector preparation: Pick several colonies from the selection
plate and incubate for 8 h to overnight at 37C in 5 mL LB media
(containing 100 mg/mL ampicillin) each. Isolate the vector
DNA using a commercial Miniprep kit (e.g., QIAprep Miniprep
kit). Make a glycerol stock of the vector by mixing 750 mL
of the remaining culture and 250 mL of sterile 60% glycerol in
a cryotube and store in 80C for later use. Verify the insert
sequence by sequencing (it is an important step because oligonucleotides often carry mutations). Once the vector sequence
is verified, prepare sufficient amount of DNA for intended
experiments.
Note: Incompatible ends in the vector allow for direct insertion of
annealed oligonucleotides without a need for dephosphorylation
of the vector and phosphorylation of oligonucleotides. However,
an incomplete restriction digest can produce a high background of empty vectors, so we recommend verifying the
presence of the insert by restriction digest prior to sequencing.
A typical cloning procedure yields up to a hundred of colonies.
If several hundred colonies and more are obtained, this usually
indicates high background of empty clones.
146
3.2.2. TMP/LMP shRNA

Vector Cloning
1. Design and obtain the sense and antisense oligonucleotides as

described in Fig. 3b.
2. Oligonucleotide annealing: Mix 5 mg of sense oligonucleotide
with 5 mg of antisense oligonucleotide in TE buffer in a total
volume of 100 mL. Place the tube with oligonucleotide mixture
in a beaker with boiling water and incubate for 2 min. Turn
heating off and leave the oligonucleotide mix in hot water to
cool down slowly. Dilute annealed oligonucleotide 100 with
TE buffer.
3. TMP vector digestion: Mix 1 mg of TMP plasmid DNA with
2 mL of 10 restriction buffer and add water to a final volume
of 18 mL. Add 1 mL of XhoI enzyme and 1 mL EcoRI enzyme
and incubate for 2 h at 37C.
4. Separate the digested plasmid in a 1.0% agarose gel in SB buffer
(sodium borate buffer: 10 mM sodium hydroxide, pH adjusted
to 8.5 with boric acid.) and extract from gel using commercial
gel extraction kit (e.g., QIAquick Gel Extraction Kit). Elute
the extracted DNA in 30 mL of elution buffer.
5. Ligation: Mix 2 mL of digested TMP DNA with 5 mL of
annealed oligonucleotide (1 ng/mL), add 2 mL of 10 ligation
buffer and nuclease-free water to a final volume of 19 mL.
Add 1 mL of T4 DNA ligase and incubate for 2 h at RT or
overnight at 16C.
6. Transformation: Add 2 mL of the ligation mixture to 50 mL
of chemically competent E. coli cells (e.g., DH5a strain) and
incubate for 20 min on ice. Submit cells to a 30 s heat-shock at
42C and immediately add 1 mL of LB medium. Incubate for
1 h at 37C with shaking.
7. Selection: Plate 100 mL of transformed cells on LB agar supplemented with 100 mg/mL ampicillin and incubate overnight
at 37C.
8. Vector preparation: Pick several colonies from the selection
plate and incubate for 8 h to overnight at 37C in 5 mL LB media
(containing 100 mg/mL ampicillin) each. Isolate the vector
DNA using a commercial Miniprep kit (e.g., QIAprep Miniprep
kit). Make a glycerol stock of the vector by mixing 750 mL of
the remaining culture and 250 mL of sterile 60% glycerol in a
cryotube and store in 80C for later use. Verify the insert
sequence by sequencing (it is an important step because oligonucleotides often carry mutations). Once verified the vector
sequence is verified, prepare sufficient amount of DNA for
intended experiments.
3.3. RNAi with

Transient Transfection
of siRNA with
Oligofectamine
Time consideration: 4 days

Before performing RNAi experiments, we recommend to test different
concentrations of each siRNA (ideally 5, 10, 20, 50, and 100 nM)
and subsequently use the lowest concentration required for
147
efficient silencing (see Note 1). The following protocol is based on

transfection of 50 nM siRNA into HEK293 and HeLa cells grown
in 6-well plates. Values/volumes indicate amounts per well, values
in parentheses indicate amounts per well for 24-well plates.
1. One day before transfection, plate cells to achieve optimal
confluency. Suggested confluency of HeLa cells for transfection
with Oligofectamine is ~50%. For HEK293 cells, we recommend
a little bit higher confluency (6070%) as they easily detach.
For other cell types, optimization of ideal confluency may be
needed. When using media with antibiotics, remove the
medium with antibiotics and replace it with a medium without
antibiotics before transfection.
2. Prepare siRNA-oligofectamine complexes. For each well, mix in a
1.5-mL tube (Tube A) 2.5 mL (1.25 mL) of siRNA (20 mM stock)
and 180 mL (190 mL) of Opti-MEM media, mix gently by tapping
on the tube, and incubate for 7 min at RT. To another tube
(tube B), add 2 mL (1 mL) of Oligofectamine and 16 mL (8 mL)
of Opti-MEM, mix gently by tapping on the tube, and incubate
for 7 min at RT. After 7-min incubation, add content of tube B
into tube A, mix gently by tapping, and incubate for 20 min at RT.
3. Transfect cells. Wash cells with Opti-MEM, add 800 mL
(400 mL) of Opti-MEM and transfection mixture from above,
and incubate for 4 h (37C). Then add 500 mL (250 mL) of
30% FCS, and incubate for 48 h. It is also possible to replace
the transfection medium with DMEM containing 10% FCS
instead of adding 30% FCS to Opti-MEM. Cells can be cultured
from 24 to 72 h after transfection but it should be kept in mind
that shorter incubation may not be sufficient for detecting
an RNAi effect and longer incubation increases the risk of
interferon response.
We also use Turbofect (Fermentas) for siRNA transfection with
good results. siRNA transfection with Turbofect is performed as
described in Subheading 3.4, except siRNA is used instead of DNA
to achieve final concentration of 40 nM during transfection.
3.4. RNAi with
Transient Transfection
of shRNA-Expressing
Vector with Turbofect
Time consideration: 1 week

Plate cells to achieve density required for transfection. Knowing
optimal density is critical for efficient transfection and may differ
from one transfection reagent to another. The following protocol
is for transfection of HEK293 and HeLa cells grown in 6-well
plates, values in parentheses indicate amounts for 24-well plates.
1. For transfection of HeLa or HEK293 cells, dilute cells in culture
medium to a density 60,000 cell/mL and plate 3 mL (0.5 mL)
of suspension per well, 24 h prior to transfection. For optimal
transfection conditions, cells should be 5070% confluent.
148
2. Change culture media just before transfection and add 1 mL

(0.5 mL) of fresh media per well.
3. Transfect cells using Turbofect transfection reagent according
to the manufacturers protocol. Briefly, combine the desired
amount of shRNA-containing plasmid (usually 1.5 mg (0.5 mg))
and pBluescript plasmid up to the total DNA amount of 2.5 mg
(1.0 mg) for each well and mix DNA with 500 mL (100 mL)
DMEM.
4. Mix by pipetting and incubate for 30 min at RT.
5. Add the mixture dropwise into each well.
6. 6 h after transfection, add 1.5 mL (0.4 mL) of fresh DMEM
medium supplemented with 10% FCS and antibiotics.
7. Harvest the cells 2472 h (typically 48 h) after transfection and
assay for RNAi effects.
Transfection efficiency of pTMP or pLMP can be easily monitored because these plasmids carry a EGFP reporter. To monitor
transfection efficiency in pSuper or pTer, one can use an EGFPexpressing plasmid for transfections instead of pBluescript.
3.5. Detection of the
Interferon Response
in RNAi Experiment
3.5.1. By RT-PCR or
Quantitative Real-Time
PCR
Table 1 lists several genes transcriptionally activated by the inteferon response found in the literature, which are suitable markers.
Particularly IFIT1 (also known as p56) seems to be a suitable, sensitive
marker for ISG activation (27, 29).
If an analysis of knockdown effects does not include RT-PCR
already, isolate total RNA from cells in 6-well plates 48 h after
transfection using Trizol (Invitrogen) according to the manufacturers
protocol. Remove DNA contamination by DNAse I treatment
(Fermentas), followed by phenol/chloroform extraction (53). Prepare
cDNA using Superscript III Reverse Transcriptase (Invitrogen)
primed with random hexamer primers according to the manufacturers instructions. Perform quantitative or semi-quantitative PCR
using the primers in Table 2.
3.5.2. In Culture
For the detection of interferon IFN-a or b by a conventional sandwich enzyme immunoassay (ELISA), some of the commercially
available kits can be used (see Table 3). The IFN concentration in
cell supernatants is measured 2448 h after transfection.
3.5.3. On Arrays
Induction of the interferon pathway in RNAi experiments is generally monitored using RT-PCR or western blotting on interferon
response marker genes or proteins, respectively. However, in experiments where effects of RNAi are studied on transcriptome level,
the interferon stimulation can be assessed directly from microarray
data. In such a case, the overall interferon stimulated gene (ISG)
profile may vary between different cell types. We explored publicly
available microarray data dealing with interferon induction in
149
Table 1
Primers for RT-PCR detection of the interferon activation
Primers for human samples
Gene name
Primer
Quantification
References
IFIT1
(p56,
ISG56-K)
F
R
CTAAGCAAAACCCTGCAGAACG
GGAATTCAATCTGATCCAAGACTC
Real-time
This work
IFIT2
F
R
GCCACAAAAAATCACAAGCCA
CCATTGTCTGGATTTAAGCGG
Real-time
(57)
RIG-I
(DDX58)
F
R
CATGTCCACCTTCAGAAGTGTCTG
GGTTTTTCCACAACCTGTAGGAGC
Real-time
This work
OAS1
F
R
CTTTGATGCCCTGGGTCAGTTG
CTCTGTAGTTCTGTGAAGCAGGTG
Real-time
This work
STAT1
F
R
TGGGTTTGACAAGGTTCTT
TATGCAGTGCCACGGAAAG
Semiquantitative
(58)
IFN-a2
F
R
GGATGAGACCCTCCTAGACAAAT
ATGATTTCTGCTCTGACAACCTC
Real-time
(59)
IFN-b
F
R
ATGAGTGGTGGTTGCAGGC
AAGCATCAGAGGCGGACTCTGGGA
Real-time
(60)
Primers for murine samples
Gene name
Primer
Quantification
References
Ifit1 (Isg56)
F
R
AGAGAGTCAAGGCAGGTTTCTGAG
TCTCACTTCCAAATCAGGTATGTCA
Real-time
This work
Ifit2 (Isg54)
F
R
ATGAAGCAGGTGCTGAATACTAGTGA
TGGTGAGGGCTTTCTTTTTCC
Real-time
(61)
Rig-I
(Ddx58)
F
R
AGCTTACTCGGAGGTTTGAAGAAA
CAGTCAGTATGCCAGGCTTTAGAA
Real-time
This work
Oas1b
F
R
AGACGTTGTGGAGTGAAGTTTGAG
T CCCAGCTTCTCCTTACACAGTTG
Semiquantitative
This work
Stat1
F
R
CACATTCACATGGGTGGAAC
TCTGGTGCTTCCTTTGGTCT
Real-time
(62)
IFN-a2
F
R
TCTGTGCTTTCCTCGTGATG
TTGAGCCTTCTGGATCTGCT
Real-time
(62)
IFN-b
F
R
GGAGATGACGGAGAAGATGC
CCCAGTGCTGGAGAAATTGT
Real-time
(63)
150
Table 2
TaqMan probes for markers for interferon activation
in human and mouse cells
Gene name
TaqMan probe
Ifit1 (ISG56)
Hs
Mm
Hs03027069_s1
Mm00515153_m1
Ifit2 (ISG54)
Hs
Mm
Hs01922738_s1
Mm00492606_m1
Rig-I
(Ddx58)
Hs
Mm
Hs01061434_m1
Mm01216860_m1
Oas1
Hs
Mm
Hs00973635_m1
Mm01198570_m1
Stat1
Hs
Mm
Hs01014000_m1
Mm01257291_m1
IFN-a2
Hs
Mm
Hs00999940_s1
Mm00833961_s1
IFN-b
Hs
Mm
Hs01077958_s1
Mm00439552_s1
Table 3
Selection of ELISA detection kits for assaying interferons
Product
Manufacturer
Reference
Human IFN-a ELISA kit
R&D Systems, USA
(64)
Human IFN-b ELISA kit
R&D Systems, USA
(64)
Mouse IFN-a ELISA kit
PBL Biomedical Laboratories,

USA
(65)
Human (mouse) IFN-b

ELISA kit
BioSource International, USA
(66)
RNAi experiments in various human cell types and generated a list

of ISGs that were induced (with one exception OAS2) in at least
two studies. These human ISGs with corresponding probe IDs for
3 widely used microarray systems are summarized in Table 4 and
their mouse counterparts in Table 5. The expression of the listed
genes should be verified in all microarray data obtained from RNAi
experiment to exclude nonspecific RNAi mediated interferon stimulation. Eight particularly important known ISG marker genes are
highlighted in bold.
RefSeq ID
NM_001111
NM_002983
NM_005195
NM_001565
NM_005409
NM_005101
NM_022873
NM_002053
NM_006877
NM_005531
NM_005532
NM_005533
NM_006417
NM_001548
NM_001547
NM_001549
NM_012420
NM_003641
NM_006435
NM_002199
NM_004029
Gene name
ADAR
CCL3
CEBPD
CXCL10
CXCL11
G1P2
G1P3
GBP1
GMPR
IFI16
IFI27
IFI35
IFI44
IFIT1
IFIT2
IFIT3
IFIT5
IFITM1
IFITM2
IRF2
IRF7
208436_s_at
203275_at
201315_x_at
214022_s_at
203595_s_at
204747_at
217502_at
203153_at
214453_s_at
209417_s_at
202411_at
206332_s_at
204187_at
231577_s_at
204415_at
205483_s_at
210163_at
204533_at
203973_s_at
205114_s_at
201786_s_at
Affymetrix
HU133 ID
ILMN_1798181
ILMN_2090607
ILMN_1673352
ILMN_1801246
ILMN_1696654
ILMN_1701789
ILMN_1739428
ILMN_1707695
ILMN_1760062
ILMN_1745374
ILMN_2058782
ILMN_1710937
ILMN_1729487
ILMN_2148785
ILMN_1687384
ILMN_2054019
ILMN_2067895
ILMN_1791759
ILMN_1782050
ILMN_1671509
ILMN_1776777
A_24_P378019
A_23_P136478
A_24_P287043
A_23_P72737
A_24_P30194
A_23_P35412
A_23_P24004
A_23_P52266
A_23_P23074
A_23_P152782
A_23_P48513
A_23_P217866
A_24_P277657
A_32_P107372
A_23_P201459
A_23_P819
A_24_P20607
A_24_P303091
A_23_P31810
A_23_P373017
A_23_P200439
(continued)
(27, 28, 67)
(24, 67)
(26, 54)
(2628, 54, 67)
(27, 54, 67)
(28, 54, 67)
(26, 28, 54)
(24, 2628, 54, 67)
(27, 54)
(27, 67)
(27, 28, 54)
(54, 67)
(27, 68)
(24, 27, 28)
(26, 28, 54)
(26, 28, 54, 67)
(27, 68)
(27, 67)
(54, 68)
(67, 68)
(24, 28, 67)
Illumina Human_WG-6v3 ID Agilent Human Genome ID References
ENSG00000185507
ENSG00000168310
ENSG00000185201
ENSG00000185885
ENSG00000152778
ENSG00000119917
ENSG00000119922
ENSG00000185745
ENSG00000137965
ENSG00000068079
ENSG00000165949
ENSG00000163565
ENSG00000137198
ENSG00000117228
ENSG00000126709
ENSG00000187608
ENSG00000169248
ENSG00000169245
ENSG00000180733
ENSG00000006075
ENSG00000160710
Ensembl ID
Table 4
List of human ISGs induced by dsRNA or in RNAi experiments inferred from published microarray data
9
151
NM_002201
NM_006084
NM_022168
NM_002462
NM_002463
NM_004688
NM_016816
NM_016817
NM_006187
NM_002759
NM_021105
NM_014314
NM_003113
NM_004509
NM_007315
NM_005419
NM_000593
NM_003265
NM_198183
NM_017414
ISG20
ISGF3G
MDA5
MX1
MX2
NMI
OAS1
OAS2
OAS3
PKR
PLSCR1
RIG-I
SP100
SP110
STAT1
STAT2
TAP1
TLR3
UBE2L6
USP18
ENSG00000184979
ENSG00000156587
ENSG00000164342
ENSG00000168394
ENSG00000170581
ENSG00000115415
ENSG00000135899
ENSG00000067066
ENSG00000107201
ENSG00000188313
ENSG00000055332
ENSG00000111331
ENSG00000111335
ENSG00000089127
ENSG00000123609
ENSG00000183486
ENSG00000157601
ENSG00000115267
ENSG00000213928
ENSG00000172183
Ensembl ID
219211_at
201649_at
206271_at
202307_s_at
205170_at
200887_s_at
209761_s_at
210218_s_at
218943_s_at
202446_s_at
204211_x_at
218400_at
204972_at
205552_s_at
203964_at
204994_at
202086_at
219209_at
203882_at
33304_at
Affymetrix
HU133 ID
ILMN_1740200
ILMN_1769520
ILMN_2155708
ILMN_1751079
ILMN_1690921
ILMN_1777325
ILMN_2415144
ILMN_2284998
ILMN_1797001
ILMN_1745242
ILMN_1706502
ILMN_2184262
ILMN_1674063
ILMN_1672606
ILMN_1739541
ILMN_2231928
ILMN_1662358
ILMN_1781373
ILMN_1745471
ILMN_1659913
A_23_P132159
A_23_P75741
A_23_P29922
A_23_P59005
A_23_P76090
A_24_P274270
A_23_P120002
A_23_P349928
A_23_P20814
A_23_P69109
A_23_P142750
A_23_P47955
A_24_P343929
A_23_P64828
A_23_P154235
A_23_P6263
A_23_P17663
A_23_P68155
A_23_P65442
A_23_P32404
(54, 67)
(26, 27)
(28, 54)
(24, 54)
(54, 67)
(26, 27, 54, 67)
(27, 54)
(27, 67)
(54, 67)
(27, 54, 67)
(54, 67)
(26, 27, 54, 67)
(26)
(2628, 67)
(24, 54, 67)
(27, 67)
(27, 28, 54)
(54, 67, 68)
(27, 54, 67)
(27, 28, 54, 67, 68)
Illumina Human_WG-6v3 ID Agilent Human Genome ID References
There are probe IDs of 3 commonly used microarray systems from Affymetrix, Illumina, and Agilent included. Eight well known ISG marker genes are
highlighted in bold
RefSeq ID
Gene name
Table 4
(continued)
152
RefSeq
NM_001038587
NM_011337
NM_007679
NM_021274
NM_019494
NM_015783
NM_010259
NM_025508
NM_008329
NM_029803
NM_027320
NM_133871
NM_008331
NM_008332
NM_010501
NM_026820
NM_030694
NM_008391
NM_016850
Gene name
Adar
Ccl3
Cebpd
Cxcl10
Cxcl11
G1p2
Gbp1
Gmpr
Ifi16
Ifi27
Ifi35
Ifi44
Ifit1
Ifit2
Ifit3
Ifitm1
Ifitm2
Irf2
Irf7
1417244_a_at
1418265_s_at
1417460_at
1424254_at
1449025_at
1418293_at
ILMN_1227573
ILMN_1251696
ILMN_1232667
ILMN_2640765
ILMN_2944666
ILMN_2981167
ILMN_2774340
ILMN_2680136
ILMN_2625290
ILMN_2762944
ILMN_3010089
ILMN_2602581
ILMN_1233293
ILMN_1256257
ILMN_1247446
ILMN_1214419
ILMN_2588570
ILMN_1253919
ILMN_2489167
Illumina MouseWG-6_v2 ID
A_51_P421876
A_51_P316523
A_51_P168459
A_52_P541802
A_51_P359570
A_52_P542388
A_51_P327751
A_51_P487690
A_51_P414889
A_52_P90363
A_51_P408343
A_51_P495986
A_51_P398766
A_52_P463936
A_52_P676403
A_51_P432641
A_51_P444447
A_51_P140710
A_52_P183181
(continued)
Agilent Mouse Genome ID
ENSMUSG00000025498
ENSMUSG00000031627
ENSMUSG00000060591
ENSMUSG00000025491
ENSMUSG00000074896
ENSMUSG00000045932
1450783_at
1423555_a_at
ENSMUSG00000028037
ENSMUSG00000034459
1445897_s_at
1426278_at
1452348_s_at
1448530_at
1420549_at
1431591_s_at
1419698_at
1418930_at
1423233_at
1419561_at
1434268_at
Affymetrix 430_v2 ID
ENSMUSG00000010358
ENSMUSG00000079017
ENSMUSG00000073489
ENSMUSG00000000253
ENSMUSG00000028269
ENSMUSG00000035692
ENSMUSG00000060183
ENSMUSG00000034855
ENSMUSG00000071637
ENSMUSG00000000982
ENSMUSG00000027951
Ensembl ID
Table 5
List of mouse counterparts to human ISGs induced in microarrays from RNAi experiments with probe IDs of 3 commonly
used microarray systems from Affymetrix, Illumina, and Agilent
9
153
NM_020583
NM_008394
NM_027835
NM_010846
NM_013606
NM_019401
NM_001083925
NM_145227
NM_145226
NM_011163
NM_011636
NM_172689
NM_013673
NM_175397
NM_009283
NM_019963
NM_013683
NM_126166
NM_019949
NM_011909
Isg20
Isgf3g
Mda5
Mx1
Mx2
Nmi
Oas1
Oas2
Oas3
Pkr
Plscr1
Rig-i
Sp100
Sp110
Stat1
Stat2
Tap1
Tlr3
Ube2l6
Usp18
ENSMUSG00000030107
ENSMUSG00000027078
ENSMUSG00000031639
ENSMUSG00000037321
ENSMUSG00000040033
ENSMUSG00000026104
ENSMUSG00000070034
ENSMUSG00000026222
ENSMUSG00000040296
ENSMUSG00000032369
ENSMUSG00000024079
ENSMUSG00000032661
ENSMUSG00000032690
ENSMUSG00000029605
ENSMUSG00000026946
ENSMUSG00000023341
ENSMUSG00000000386
ENSMUSG00000026896
ENSMUSG00000002325
ENSMUSG00000039236
Ensembl ID
Eight well-known ISG marker genes are highlighted in bold
RefSeq
Gene name
Table 5
(continued)
1418191_at
1417172_at
1422782_s_at
1416016_at
1421911_at
1420915_at
1456493_at
1451821_a_at
1456890_at
1429527_a_at
1440866_at
1425374_at
1425065_at
1425119_at
1425719_a_at
1419676_at
1451905_a_at
1426276_at
1421322_a_at
1419569_a_at
Affymetrix 430_v2 ID
ILMN_2433990
ILMN_2431619
ILMN_2697002
ILMN_1250409
ILMN_2657822
ILMN_2510233
ILMN_1214911
ILMN_2846812
ILMN_2717127
ILMN_2911344
ILMN_1250410
ILMN_1216020
ILMN_2670150
ILMN_2613140
ILMN_2755958
ILMN_1239219
ILMN_2707870
ILMN_2648913
ILMN_1233461
ILMN_2735615
Illumina MouseWG-6_v2 ID
A_51_P164219
A_52_P214740
A_52_P85174
A_51_P100327
A_51_P225808
A_52_P496503
A_52_P512201
A_52_P127720
A_52_P523946
A_52_P654841
A_52_P559919
A_51_P472867
A_52_P587071
A_52_P110877
A_51_P125067
A_51_P514085
A_52_P446431
A_52_P121468
A_52_P176013
A_51_P510713
Agilent Mouse Genome ID
154
155
Table 6
siRNA sequences and oligos for shRNA expression
for deliberate induction of interferon response
by small RNAs
3.6. Deliberate
Induction of
Interferon Response
3.6.1. With Poly I:C
3.6.2. With InterferonInducing shRNA/siRNAs
siRNA (sense strand)
Reference
GUCCGGGCAGGUCUACUUUTT
(36)
AGCUUAACCUGUCCUUCAAdTdT
(35)
UGUCCUUCAAUGUCCUUCAA
(35)
CUACACAAAUCAGCGAUUU
(34)
shRNA (designed for TMP vector)
Reference
TGCTGTTGACAGTGAGCGCTACACAAATCAGCG
ATTTTTTTAGTGAAGCCACAGATGTAAAAAAAT
CGCTGATTTGTGTAGTGCCTACTGCCTCGGA
This work
TGCTGTTGACAGTGAGCGTGTCCTTCAATGTCCT
TCAATTTAGTGAAGCCACAGATGTAAATTGAAGG
ACATTGAAGGACATGCCTACTGCCTCGGA
This work
The production of interferons in a positive control sample can be

induced by direct addition of the polyinosinic:polycytidylic acid
(polyI:C) to the cultured cells. The suggested final concentration
in culture media is 50 mg/mL of polyI:C. Add polyI:C to cells 6 h
after transfection.
As discussed above, some siRNAs may stimulate the interferon
response in immune cells (see Table 6). Some siRNAs contain specific immunostimulatory motifs within single-stranded RNAs
(ssRNA) from siRNAs, such as UGUGU and GUCCUUCAA,
(34, 35), while others do not have defined sequence motifs (36).
As positive control for interferon induction, we use two of these
siRNAs expressed from pTMP vector (Fig. 4). Oligonucleotides
used for production of pTMP vectors are listed in Table 6. For
transfection, use the same protocol as above.
4. Notes
A good experiment should include appropriate controls, which
would help to pinpoint the cause of a problem. The most common
problems and their solutions are discussed below.
1. Troubleshooting poor RNAi knockdown. It can be caused by
problems with the vector, its delivery, or inefficient siRNA.
156
Fig. 4. Interferon response to immunostimulatory shRNA expressed from pTMP vectors.

(a) EGFP expression from pTMP EGFP reporter in HeLa cells transfected with pTMP vectors expressing immunostimulatory shRNA (0.5 mg per well in a 6-well plate). (b) Induction
of interferon response markers by expressing immunostimulatory shRNAs. HeLa cells
were transfected with 0.5 mg of each plasmid pTMP plasmid derivate per well in a 6-well
plate and relative interferon response marker expression was estimated by quantitative
Real-Time RT-PCR. Expression of the markers in cells transfected with empty pTMP vector
was set to 1.
Check the quality of the vector DNA (electrophoresis and

sequencing).
Include a positive transfection control.
Increase the amount of the shRNA vector in transfection.
Assay at different time points (protein is downregulated later

than mRNA).
Try another siRNA sequence.

2. Handling interferon response. If the interferon response is
detected in an RNAi experiment, the following troubleshooting procedure may help. Verify that the RNAi trigger and not
the delivery or induction method cause the effect since delivery methods themselves may induce the interferon response.
157
A typical experiment should include appropriate controls, which

would help to identify the cause. Changing the amount of the
transfection reagent, using another transfection reagent,
producing stable inducible cell lines, and titration of inducing
and selecting agents to lower doses may help if the interferon
response is caused by other factors than the RNAi trigger itself.
If the interferon response is detected in an RNAi experiment, the following troubleshooting procedure may help.
Verify that the RNAi trigger and not the delivery or induction
method cause the effect since delivery methods themselves
may induce the interferon response. A typical experiment
should include appropriate controls, which would help to identify the cause. Changing the amount of the transfection reagent,
using another transfection reagent, producing stable inducible
cell lines, and titration of inducing and selecting agents to
lower doses may help if the interferon response is caused by
other factors than the RNAi trigger itself.
If the RNAi trigger seems to be causing the effect and it is
an siRNA, the following solutions may work. If the siRNA is
from a commercial source, titrate the amount of the siRNA to
the lowest effective dose first and/or use an earlier time-point
for analysis. For example, the interferon response may be
observed at 72 h after transfection but not at 24 or 48 h posttransfection (28). If one needs rather late time-points, it is
worth considering the production of inducible stable cell lines.
The purity of the siRNA is also very important and changing
the manufacturer may help in some cases. Poor quality of
chemically synthesized siRNAs may induce the interferon
response (29). Also, avoid motifs that stimulate the interferon
response discussed above (see Subheading 1.3). Sometimes,
switching to another cell line can help, as different cell lines
have different sensitivity to the interferon activation (54). As
mentioned above, if the siRNAs are home-made by T7 polymerase, treatment of siRNAs with RNAse T1 and alkaline
phosphatase to remove 5 triphosphate GTP should help (30).
If the problem still persists, it is probably more economical to
purchase siRNAs from one of the established sources rather
than spending time and financial resources to solve the problem. If the problem appears to be linked to an shRNA expressed
from a vector controlled by the U6 or H1 promoter, it is advisable to verify that the vector sequence does not contain the
critical AA dinucleotide motif near the transcription start site
(33). If that is not the case, one should try using another
siRNA sequence and/or expression vector (e.g., switching
from the type I to the type II hairpin).
3. Dealing with off-targeting. A commentary by Echeverri et al.
(42) summarizing the guidelines for appropriate controls in
RNAi experiments offers two ways to address off-target effects in
RNAi experiments named the two Rs: rescue or redundancy.
158
Rescue experiments are performed by delivering expression

of a functional version of a targeted gene, which is mutated
such that the base-pairing with a short RNA is disrupted. If the
short RNA is targeting the 3UTR, it should not be a problem
to appropriately replace the 3UTR. If the short RNA is targeting the coding sequence, the situation is little bit more complicated because one has to mutate/degenerate numerous
positions within codons of the target sequence. It is important
to consider codon usage so as not to introduce more rare
codons than necessary. One or two nucleotides should be
mutated in the middle of the sequence to impair Ago2mediated cleavage. In addition, it is useful to mutate one
nucleotide in a position complementary to nucleotides 24 of
the binding siRNA sequence to interfere with recognition of
the sequence by RISC.
Redundancy is another way to address off-target effects.
Two or more RNA triggers with distinct sequences producing
the same phenotype decrease the probability that the phenotype is caused by off-targeting. However, some phenotypes are
fairly common (e.g., slower growth, apoptosis, developmental
arrest), so using two or even more different RNAi inducers may
not be enough to decipher if the phenotype is specific to the
downregulated gene or not. In such a case, a rescue experiment
is highly advisable. When microarray analysis is being performed
on knockdown cells, one can estimate the extent of off-targeting on the cellular transcriptome for each short RNA used.
Similarly to analysis of hexamer and heptamer seed enrichment
among upregulated transcripts upon repression of RNA silencing (55), one can test if the seeding motif corresponding to the
5 end of an active siRNA strand is significantly enriched in
3UTRs of transcripts downregulated in knockdown cells.
It must be pointed out that nontargeting controls
(scrambled siRNAs or shRNAs, or short RNAs against nonexpressed genes such as EGFP, RL-luciferase, etc.) are not appropriate controls for off-targeting for reasons mentioned above.
It is a common misconception that ignores the fact that offtargeting is individual to each RNAi trigger because it is
sequence-specific. Nontargeting RNAi triggers rather serve
as controls for the sequence-independent effects, such as interferon response and saturation of RNA silencing with an excess
of exogenous short RNAs.
Acknowledgements
We thank Witold Filipowicz group at the FMI for sharing their
experience and protocols and Daniela Schmitter, Radek Malik, and
Lenka Sarnova for help with preparation of the manuscript.
159
Protocols are based on research supported by the GAAV grant IAA

501110701, EMBO SDIG grant 20061483, GACR grant
204/09/0085, and the Purkynje Fellowship.
References
1. Sontheimer, E. J., Carthew, R. W. (2005)
Silence from within: endogenous siRNAs and
miRNAs. Cell 122: 912.
2. Zamore, P. D., Haley, B. (2005) Ribo-gnome:
the big world of small RNAs. Science 309:
15191524.
3. Filipowicz, W. (2005) RNAi: the nuts and bolts
of the RISC machine. Cell 122: 1720.
4. Tolia, N. H., Joshua-Tor, L. (2007) Slicer and
the argonautes. Nat. Chem. Biol. 3: 3643.
5. Elbashir, S. M., Harborth, J., Lendeckel, W.,
Yalcin, A., Weber, K., Tuschl, T. (2001)
Duplexes of 21-nucleotide RNAs mediate RNA
interference in cultured mammalian cells.
Nature 411: 494498.
6. Manche, L., Green, S. R., Schmedt, C., Mathews,
M. B. (1992) Interactions between doublestranded RNA regulators and the protein kinase
DAI. Mol. Cell. Biol. 12: 52385248.
7. Svoboda, P. (2008) RNA silencing in mammalian oocytes and early embryos. Curr. Top.
Microbiol. Immunol. 320: 225256.
8. Mineno, J., Okamoto, S., Ando, T., Sato, M.,
Chono, H., Izu, H., Takayama, M., Asada, K.,
Mirochnitchenko, O., Inouye, M., Kato, I.
(2006) The expression profile of microRNAs in
mouse embryos. Nucleic Acids Res. 34:
17651771.
9. Cummins, J. M., He, Y., Leary, R. J., Pagliarini,
R., Diaz, L. A., Jr., Sjoblom, T., Barad, O.,
Bentwich, Z., Szafranska, A. E., Labourier, E.,
Raymond, C. K., Roberts, B. S., Juhl, H.,
Kinzler, K. W., Vogelstein, B., Velculescu, V. E.
(2006) The colorectal microRNAome. Proc.
10. Liu, J., Carmell, M. A., Rivas, F. V., Marsden,
C. G., Thomson, J. M., Song, J. J., Hammond,
S. M., Joshua-Tor, L., Hannon, G. J. (2004)
Argonaute2 is the catalytic engine of mammalian RNAi. Science 305: 14371441.
11. Meister, G., Landthaler, M., Patkaniowska, A.,
Dorsett, Y., Teng, G., Tuschl, T. (2004)
Human Argonaute2 mediates RNA cleavage
targeted by miRNAs and siRNAs. Mol. Cells
15: 185197.
12. Pillai, R. S., Artus, C. G., Filipowicz, W. (2004)
Tethering of human AGO proteins to mRNA
mimics the miRNA-mediated repression of
protein synthesis. RNA - Publ. RNA Soc. 10:
15181525.
13. Song, J. J., Smith, S. K., Hannon, G. J., JoshuaTor, L. (2004) Crystal structure of Argonaute
and its implications for RISC slicer activity.
Science 305: 14341437.
14. Doench, J. G., Petersen, C. P., Sharp, P. A.
(2003) siRNAs can function as miRNAs. Genes
Dev. 17: 438442.
15. Hutvagner, G., Zamore, P. D. (2002) A
microRNA in a multiple-turnover RNAi
enzyme complex. Science 297: 20562060.
16. Yekta, S., Shih, I. H., Bartel, D. P. (2004)
MicroRNA-directed cleavage of HOXB8
mRNA. Science 304: 594596.
17. Brummelkamp, T. R., Bernards, R., Agami, R.
(2002) A system for stable expression of short
interfering RNAs in mammalian cells. Science
296: 550553.
18. McManus, M. T., Petersen, C. P., Haines, B.
B., Chen, J., Sharp, P. A. (2002) Gene silencing using micro-RNA designed hairpins. RNA
- Publ. RNA Soc. 8: 842850.
19. Paddison, P. J., Caudy, A. A., Bernstein, E.,
Hannon, G. J., Conklin, D. S. (2002) Short
hairpin RNAs (shRNAs) induce sequence-specific silencing in mammalian cells. Genes Dev.
16: 948958.
20. van de Wetering, M., Oving, I., Muncan, V.,
Pon Fong, M. T., Brantjes, H., van Leenen, D.,
Holstege, F. C., Brummelkamp, T. R., Agami,
R., Clevers, H. (2003) Specific inhibition of
gene expression using a stably integrated,
inducible small-interfering-RNA vector. EMBO
Rep. 4: 609615.
21. Chung, K. H., Hart, C. C., Al-Bassam, S.,
Avery, A., Taylor, J., Patel, P. D., Vojtek, A. B.,
Turner, D. L. (2006) Polycistronic RNA polymerase II expression vectors for RNA interference based on BIC/miR-155. Nucleic Acids
Res. 34: e53.
22. Zeng, Y., Wagner, E. J., Cullen, B. R. (2002) Both
natural and designed micro RNAs can inhibit the
expression of cognate mRNAs when expressed in
human cells. Mol. Cells 9: 13271333.
23. de Veer, M. J., Sledz, C. A., Williams, B. R.
(2005) Detection of foreign RNA: implications
for RNAi. Immunol. Cell Biol. 83: 224228.
24. Geiss, G., Jin, G., Guo, J., Bumgarner, R.,
Katze, M. G., Sen, G. C. (2001) A comprehensive view of regulation of gene expression by
160
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
double-stranded RNA-mediated cell signaling. J.
Biol. Chem. 276: 3017830182.
Schlee, M., Hornung, V., Hartmann, G. (2006)
siRNA and isRNA: two edges of one sword.
Mol. Ther. 14: 463470.
Sledz, C. A., Holko, M., de Veer, M. J.,
Silverman, R. H., Williams, B. R. (2003)
Activation of the interferon system by shortinterfering RNAs. Nat. Cell Biol. 5: 834839.
Bridge, A. J., Pebernard, S., Ducraux, A.,
Nicoulaz, A. L., Iggo, R. (2003) Induction of
an interferon response by RNAi vectors in mammalian cells. Nature Genet. 34: 263264.
Svoboda, P. (2007) Off-targeting and other nonspecific effects of RNAi experiments in mammalian cells. Curr. Opin. Mol. Ther. 9: 248257.
Marques, J. T., Devosse, T., Wang, D.,
Zamanian-Daryoush, M., Serbinowski, P.,
Hartmann, R., Fujita, T., Behlke, M. A.,
Williams, B. R. (2006) A structural basis for
discriminating between self and nonself double-stranded RNAs in mammalian cells. Nat.
Biotechnol. 24: 559565.
Kim, D. H., Longo, M., Han, Y., Lundberg, P.,
Cantin, E., Rossi, J. J. (2004) Interferon induction by siRNAs and ssRNAs synthesized by
phage polymerase. Nat. Biotechnol. 22:
321325.
Hornung, V., Ellegast, J., Kim, S., Brzozka, K.,
Jung, A., Kato, H., Poeck, H., Akira, S.,
Conzelmann, K. K., Schlee, M., Endres, S.,
Hartmann, G. (2006) 5-Triphosphate RNA is
the ligand for RIG-I. Science 314: 994997.
Pichlmair, A., Schulz, O., Tan, C. P., Naslund,
T. I., Liljestrom, P., Weber, F., Reis e Sousa, C.
(2006) RIG-I-mediated antiviral responses to
single-stranded RNA bearing 5-phosphates.
Science 314: 9971001.
Pebernard, S., Iggo, R. D. (2004) Determinants
of interferon-stimulated gene induction by
RNAi vectors. Differentiation 72: 103111.
Judge, A. D., Sood, V., Shaw, J. R., Fang, D.,
McClintock, K., MacLachlan, I. (2005)
Sequence-dependent stimulation of the mammalian innate immune response by synthetic
siRNA. Nat. Biotechnol. 23: 457462.
Hornung, V., Guenthner-Biller, M., Bourquin,
C., Ablasser, A., Schlee, M., Uematsu, S.,
Noronha, A., Manoharan, M., Akira, S., de
Fougerolles, A., Endres, S., Hartmann, G.
(2005) Sequence-specific potent induction of
IFN-alpha by short interfering RNA in plasmacytoid dendritic cells through TLR7. Nat. Med.
11: 263270.
Sioud, M. (2005) Induction of inflammatory
cytokines and interferon responses by doublestranded and single-stranded siRNAs is
sequence-dependent and requires endosomal
localization. J. Mol. Biol. 348: 10791090.
37. Marques, J. T., Williams, B. R. (2005)

Activation of the mammalian immune system
by siRNAs. Nat. Biotechnol. 23: 13991405.
38. Jackson, A. L., Bartz, S. R., Schelter, J.,
Kobayashi, S. V., Burchard, J., Mao, M., Li, B.,
Cavet, G., Linsley, P. S. (2003) Expression profiling reveals off-target gene regulation by
RNAi. Nat. Biotechnol. 21: 635637.
39. Jackson, A. L., Burchard, J., Schelter, J., Chau,
B. N., Cleary, M., Lim, L., Linsley, P. S. (2006)
Widespread siRNA off-target transcript
silencing mediated by seed region sequence
complementarity. RNA - Publ. RNA Soc. 12:
11791187.
40. Scacheri, P. C., Rozenblatt-Rosen, O., Caplen,
N. J., Wolfsberg, T. G., Umayam, L., Lee, J. C.,
Hughes, C. M., Shanmugam, K. S.,
Bhattacharjee, A., Meyerson, M., Collins, F. S.
(2004) Short interfering RNAs can induce
unexpected and divergent changes in the levels
of untargeted proteins in mammalian cells. Proc.
41. Brown, K., Samarsky, D. (2005) RNAi off-targeting: Light at the end of the tunnel. J. RNAi
Gene Silencing 2: 175177.
42. Echeverri, C. J., Beachy, P. A., Baum, B., Boutros,
M., Buchholz, F., Chanda, S. K., Downward, J.,
Ellenberg, J., Fraser, A. G., Hacohen, N., Hahn,
W. C., Jackson, A. L., Kiger, A., Linsley, P. S.,
Lum, L., Ma, Y., Mathey-Prevot, B., Root, D. E.,
Sabatini, D. M., Taipale, J., Perrimon, N.,
Bernards, R. (2006) Minimizing the risk of
reporting false positives in large-scale RNAi
screens. Nat. Methods 3: 777779.
43. Snove, O., Jr., Holen, T. (2004) Many commonly used siRNAs risk off-target activity.
Biochem. Biophys. Res. Commun. 319: 256263.
44. Jackson, A. L., Burchard, J., Leake, D.,
Reynolds, A., Schelter, J., Guo, J., Johnson, J.
M., Lim, L., Karpilow, J., Nichols, K., Marshall,
W., Khvorova, A., Linsley, P. S. (2006) Positionspecific chemical modification of siRNAs
reduces off-target transcript silencing. RNA
- Publ. RNA Soc. 12: 11971205.
45. Birmingham, A., Anderson, E., Sullivan, K.,
Reynolds, A., Boese, Q., Leake, D., Karpilow,
J., Khvorova, A. (2007) A protocol for designing siRNAs with high functionality and specificity. Nat. Protoc. 2: 20682078.
46. Peek, A. S., Behlke, M. A. (2007) Design of
active small interfering RNAs. Curr. Opin. Mol.
Ther. 9: 110118.
47. Huesken, D., Lange, J., Mickanin, C., Weiler,
J., Asselbergs, F., Warner, J., Meloon, B.,
Engel, S., Rosenberg, A., Cohen, D., Labow,
M., Reinhardt, M., Natt, F., Hall, J. (2005)
Design of a genome-wide siRNA library using
an artificial neural network. Nat. Biotechnol.
23: 9951001.
48. Tafer, H., Ameres, S. L., Obernosterer, G.,

Gebeshuber, C. A., Schroeder, R., Martinez, J.,
Hofacker, I. L. (2008) The impact of target site
accessibility on the design of effective siRNAs.
Nat. Biotechnol. 26: 578583.
49. Chalk, A. M., Sonnhammer, E. L. L. (2008)
siRNA specificity searching incorporating
mismatch tolerance data. Bioinformatics 24:
13161317.
50. Matveeva, O., Nechipurenko, Y., Rossi, L.,
Moore, B., Saetrom, P., Ogurtsov, A. Y., Atkins,
J. F., Shabalina, S. A. (2007) Comparison of
approaches for rational siRNA design leading
to a new efficient and transparent method.
Nucleic Acids Res. 35: e63.
51. Ameres, S. L., Martinez, J., Schroeder, R.
(2007) Molecular basis for target RNA recognition and cleavage by human RISC. Cell 130:
101112.
52. Vert, J. P., Foveau, N., Lajaunie, C.,
Vandenbrouck, Y. (2006) An accurate and
interpretable model for siRNA efficacy prediction. BMC Bioinformatics 7: 520.
53. Sambrook, J., Fritsch, E. F., Maniatis, T. (1989)
Molecular cloning: a laboratory manual. 2nd
ed., Cold Spring Harbor Laboratory, Cold
Spring Harbor, N.Y.
54. Reynolds, A., Anderson, E. M., Vermeulen, A.,
Fedorov, Y., Robinson, K., Leake, D., Karpilow,
J., Marshall, W. S., Khvorova, A. (2006)
Induction of the interferon response by siRNA
is cell type- and duplex length-dependent.
RNA - Publ. RNA Soc.12: 988993.
55. Schmitter, D., Filkowski, J., Sewer, A., Pillai, R.
S., Oakeley, E. J., Zavolan, M., Svoboda, P.,
Filipowicz, W. (2006) Effects of Dicer and
Argonaute down-regulation on mRNA levels
in human HEK293 cells. Nucleic Acids Res.
34: 48014815.
56. Zeng, Y., Cai, X., Cullen, B. R. (2005) Use of
RNA polymerase II to transcribe artificial
microRNAs. Methods Enzymol. 392: 371380.
57. Abe, Y., Sinozaki, H., Takagi, T., Minegishi, T.,
Kokame, K., Kangawa, K., Uesaka, M.,
Miyamoto, K. (2006) Identification of
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)inducible genes in human amniotic epithelial
cells. Reprod. Biol. Endocrinol. 4: 27.
58. Hu, X., Herrero, C., Li, W. P., Antoniv, T. T.,
Falck-Pedersen, E., Koch, A. E., Woods, J. M.,
Haines, G. K., Ivashkiv, L. B. (2002) Sensitization
of IFN-gamma Jak-STAT signaling during macrophage activation. Nat. Immunol. 3: 859866.
59. Kato, A., Homma, T., Batchelor, J., Hashimoto,
N., Imai, S., Wakiguchi, H., Saito, H.,
161
Matsumoto, K. (2003) Interferon-alpha/beta

receptor-mediated selective induction of a gene
cluster by CpG oligodeoxynucleotide 2006.
BMC Immunol. 4: 8.
60. Kamezaki, K., Shimoda, K., Numata, A.,
Matsuda, T., Nakayama, K., Harada, M. (2004)
The role of Tyk2, Stat1 and Stat4 in LPSinduced endotoxin signals. Int. Immunol. 16:
11731179.
61. Sakamoto, S., Qin, J., Navarro, A., Gamero,
A., Potla, R., Yi, T., Zhu, W., Baker, D. P.,
Feldman, G., Larner, A. C. (2004) Cells
Previously Desensitized to Type 1 Interferons
Display Different Mechanisms of Activation of
Stat-dependent Gene Expression from Naive
Cells. J. Biol. Chem. 279: 32453253.
62. Wang, Z. W., Sarmento, L., Wang, Y., Li,
X.-q., Dhingra, V., Tseggai, T., Jiang, B., Fu,
Z. F. (2005) Attenuated Rabies Virus
Activates, while Pathogenic Rabies Virus
Evades, the Host Innate Immune Responses
in the Central Nervous System. J. Virol. 79:
1255412565.
63. Stewart, M. J., Smoak, K., Blum, M. A., Sherry,
B. (2005) Basal and Reovirus-Induced Beta
Interferon (IFN-{beta}) and IFN-{beta}Stimulated Gene Expression Are Cell Type
Specific in the Cardiac Protective Response.
J. Virol. 79: 29792987.
64. Pauls, E., Senserrich, J., Bofill, M., Clotet, B.,
Este, J. A. (2007) Induction of interleukins
IL-6 and IL-8 by siRNA. Clin. Exp. Immunol.
147: 189196.
65. Yang, R., Murillo, F. M., Cui, H., Blosser, R.,
Uematsu, S., Takeda, K., Akira, S., Viscidi, R.
P., Roden, R. B. S. (2004) Papillomavirus-Like
Particles Stimulate Murine Bone MarrowDerived Dendritic Cells To Produce Alpha
Interferon and Th1 Immune Responses via
MyD88. J. Virol. 78: 1115211160.
66. Bolcato-Bellemin, A.-L., Bonnet, M.-E.,
Creusat, G., Erbacher, P., Behr, J.-P. (2007)
Sticky overhangs enhance siRNA-mediated
gene silencing. Proc. Natl. Acad. Sci. USA 104:
1605016055.
67. Cekaite, L., Furset, G., Hovig, E., Sioud, M.
(2007) Gene expression analysis in blood cells
in response to unmodified and 2-modified siRNAs reveals TLR-dependent and independent
effects. J. Mol. Biol. 365: 90108.
68. Persengiev, S. P., Zhu, X., Green, M. R. (2004)
Nonspecific, concentration-dependent stimulation
and repression of mammalian gene expression
by small interfering RNAs (siRNAs). RNA Publ. RNA Soc. 10: 1218.
Chapter 10
shRNA-Induced Interferon-Stimulated Gene Analysis
Nria Morral and Scott R. Witting
Abstract
RNA interference (RNAi) is a cellular mechanism to inhibit the expression of gene products in a highly
specific manner. In recent years, RNAi has become the cornerstone of gene function studies, shortening
the otherwise long process of target identification and validation. In addition, small interfering RNA
(siRNA) and short-hairpin RNA (shRNA) therapies are being developed for the treatment of a variety of
human diseases. Despite its huge potential for gene silencing, a hurdle to safe and effective RNAi is the
activation of innate immune responses. Induction of innate immunity is dose- and sequence-dependent,
and is also influenced by target tissue and delivery vehicle. Research on the molecular mechanisms mediating this response is helping to improve the design of the RNAi molecules. Nevertheless, appropriate
testing for the presence of this undesired effect is needed prior to making conclusions on the outcome
of the silencing treatment.
Key words: RNA interference, Short-hairpin RNA, Gene transfer, Animal models, Interferon
response, Interferon-stimulated gene
1. Introduction
RNA interference (RNAi) is the phenomenon by which noncoding
double-stranded RNA (dsRNA) suppresses the expression of a
gene. Although initially discovered in the nematode Caenorhabditis
elegans, it was later realized that this is a widespread system to
regulate gene expression, and an important mechanism to coordinate complex developmental as well as physiological processes in
higher eukaryotes. RNAi is mediated by short RNA molecules of
2128 nucleotides (nt) that bind to target mRNA, triggering protein translation inhibition or mRNA degradation, based on the
degree of homology between the dsRNA molecule and the mRNA.
RNAi has become an exceptional tool in molecular biology
163
164
N. Morral and S.R. Witting
research, shortening the time-consuming process of studying gene

function and generating novel drug treatments. In addition, this
technology is being used as a therapeutic drug to suppress gene
expression. Small interfering RNA (siRNA) are chemically synthesized dsRNA molecules that upon delivery into cells or tissues,
block gene expression for a few days. Re-administration is necessary for prolonged silencing. Alternatively, short-hairpin RNA
(shRNA) can be used to provide a continuous source of silencing
molecules. shRNA expression cassettes are engineered using RNA
polymerase II or RNA polymerase III promoters, and can be
delivered to cells using plasmids or viral vectors such as retrovirus,
lentivirus, adenovirus, or adeno-associated virus (AAV).
An undesirable aspect of RNAi that limits its use is the activation of innate immunity. Mammalian cells have evolved to recognize
single or double-stranded RNA as part of the innate immune
response to pathogen infection, and siRNA/shRNA have the
capacity to activate similar responses. Cells respond to the presence
of foreign RNA by activating RNA binding proteins such as
2,5-oligoadenylate synthetases (OASs), dsRNA-dependent
protein kinase (PKR), cell surface and endosome-expressed Tolllike receptors (TLRs) 3, 7, and 8, and the cytosolic ssRNA/dsRNA
sensors retinoic acid-inducible gene I (RIG-I) and melanoma
differentiation-associated gene 5 (MDA5) (1, 2). The RNA sensor
2,5-oligoadenylate synthetase promotes the conversion of ATP
into short 2,5-oligoadenylates that bind to and activate RNaseL,
resulting in cleavage of cellular and viral RNA, and inhibition of
protein translation (3, 4). PKR, RIG-I, MDA5, and TLR activation
initiate signal transduction events that lead to the production of interferons (IFNs) (58), and consequent up-regulation of interferonstimulated genes (ISG). Interferons represent part of the innate
immune system and can induce proapoptotic events in response to
bacteria, virus, and fungi infections to protect other cells from
infection (9). While TLR activation is triggered predominantly in
immune cells, cytoplasmic RNA sensors such as PKR and RIG-I
are present in nonimmune cells and can induce type I IFN (IFNa/b) and inflammatory cytokine production (10, 11). Activation
of the IFN response is sequence-dependent. For example, the
GUCCUUCAA motif activates TLR7 (12), and the UGUGU
motif is a potent inducer of the interferon response (13). More
recently, GU-rich 4-mer RNA sequences such as UUGU, GUUC,
GUUU, UUUC, UGUU, or UCUC, have been shown to stimulate
TLR7 and TLR8 activity and produce IFN-a (14, 15).
Soon after the discovery of RNAi, siRNA of 2123 nt were
found to bypass PKR activation and subsequent IFN production
(16), triggering a major interest in using the system to silence gene
expression in vitro (1619) and in vivo (2027). However, it has
become clear that siRNA and shRNA molecules have the capacity
to induce these unspecific effects (7, 13, 26, 2831) through
10
165
ssRNA or dsRNA recognition by RIG-1, MDA-5, and TLRs (28,

32, 33). siRNA activate IFN production in a sequence and dosedependent manner (12, 13, 26). Interestingly, incorporating
2-O-methyl (2OMe) uridine or guanosine nucleosides into the
antisense strand of siRNA prevented an interferon response (34).
Other structures that have been found associated with interferon
production are blunt-ended dsRNA and ssRNA with 5-triphosphates
(3537). We have found that adenovirus-mediated expression of
shRNA in liver results in similar responses to those observed with
siRNA, i.e., activation of IFN production is sequence and dosedependent (26, 27). An important aspect of in vivo studies is the
fact that expression of the shRNA is likely to occur in immune cells
present in the target tissue. For example, the liver is a complex
organ with multiple cell types (38). Hepatocytes comprise approximately 60% of cells, while sinusoidal endothelial cells (SECs),
Kupffer cells, and hepatic stellate cells represent 20, 15, and 5% of
cells, respectively (38). Adenoviral vectors transduce hepatocytes
very efficiently (39, 40), and can infect hepatic stellate cells (41),
and macrophages (42, 43). Kupffer cells are specialized macrophages
located in the walls of the sinusoids and form part of the mononuclear
phagocyte system. Thus, these cells have high potential for recognizing shRNA hairpin structures as foreign RNAs and stimulate
interferon production.
Based on these observations, it has become essential that investigators conduct rigorous studies to determine whether innate
immunity is induced in their specific gene silencing application.
One of the most sensitive ways to measure the presence of immunostimulation is quantification of mRNA levels of known interferonstimulated genes, such as OAS (1), interferon-stimulated gene 56
(ISG56, also known as IFIT1), and interferon-stimulated exonuclease gene 20 kDa (ISG20) (44). The measurement of secreted
cytokines is not the most reliable method, as cytokine levels increase
for a short time and may not be detectable systemically. Thus, a
negative result does not necessarily mean absence of induction.
Gene expression quantification in the tissue of interest remains the
most reliable and sensitive assay. In addition, tissue sections can be
evaluated histologically to look for signs of inflammation and cell
death, to confirm the gene expression data.
2. Materials
2.1. shRNA Expression
Cassette
1. BLOCK-iT U6 RNAi Entry Vector Kit (Invitrogen,

Carlsbad, CA).
2. Top and bottom strand oligonucleotide primers.
3. E. coli XL1-Blue Subcloning-Grade
(Stratagene, La Jolla, CA).
Competent
Cells
166
4. LB powder (Sigma, St. Louis, MO). Dissolve 25 g in 1 L

deionized water. Autoclave for 15 min at 121C. Store at room
temperature.
5. Ampicillin: 100 mg/mL. Store at 20C.
6. LB agar EZmix (Sigma) ampicillin (50 mg/mL) plates.
Suspend 35.6 g in 1 L of deionized water. Autoclave for 15 min
at 121C. Transfer to a water bath at 50C for 30 min. Add
0.5 mL 100 mg/mL ampicillin. Mix well and distribute
3040 mL per plate. Store at 4C.
7. QIAprep Spin Miniprep Kit (Qiagen, Valencia, CA).
8. SeaKem LE Agarose (Lonza, Rockland, ME).
9. 10 DNA gel loading buffer (5 Prime, Gaithersburg, MD).
10. TBE solution (5): 270 g Tris base, 137.5 g boric acid, 100 mL
0.5 M EDTA, pH 8. Store at room temperature. Working 1
solutions are prepared by dilution in deionized water.
11. Molecular weight DNA marker: 1 kb ladder (New England
Biolabs, Ipswich, MA).
2.2. Cell Culture
1. HEK293 cells (Microbix Biosystems, Toronto, Ontario).

2. Dulbeccos Modified Eagles Medium (DMEM) (Gibco/BRL,
Bethesda, MD) supplemented with 10% (v/v) fetal bovine
serum (FBS, HyClone, Ogden, UT) and 100 U/mL penicillin,
100 mg/mL streptomycin (Gibco/BRL).
3. Trypsin/EDTA: 0.05% trypsin, 0.53 mM EDTA (Mediatech
Inc., Manassas, VA).
4. Metafectene Pro (Biontex-USA, San Diego, CA).
2.3. RNA Isolation
1. RNeasy Maxi kit (Qiagen).

2. Nuclease-free water (Ambion, Austin, TX).
2.4. Real-Time RT-PCR
1. QuantiTect SYBR Green RT-PCR kit, real-time One-Step

RT-PCR (Qiagen, Valencia, CA).
2. ABI 7500 real-time PCR System (Applied Biosystems, Foster
City, CA).
3. MicroAmp Optical 96-Well Reaction Plate with Barcode
(Applied Biosystems).
4. MicroAmp Optical 8-cap strip (Applied Biosystems).
5. Primers: forward and reverse (50 nmol scale). Prepare a stock
solution of 100 pmol/mL, in 10 mM TrisHCl, pH 7.5
(Ambion). Store at 20C.
2.5. Liver Enzyme

Function Markers
Alanine Aminotransferase (ALT) kit (Pointe Scientific, Canton, MI).
10
167
3. Methods
3.1. Hairpin Design
Two features are important to assemble an shRNA expression

cassette: (1) promoter and (2) sequence of the hairpin. RNA polymerase II (45, 46) and RNA polymerase III (17, 18, 47) promoters
have been used to drive expression of shRNA. The fact that cellular
microRNA are largely transcribed from RNA polymerase II promoters, opened the possibility to use this type of promoter for expression
of shRNA. However, transcripts generated by RNA polymerase II
include 5 end caps and 3 polyadenylated [poly(A)] tails, which
could potentially interfere with processing of the hairpin. One successful approach has been to design a minimal poly(A) signal (22). The
cytomegalovirus (CMV) promoter has been used to drive expression
of an shRNA embedded in the well-characterized miR30 transcript
followed by a minimal polyA signal (4850). The RNA polymerase III
promoters of the small nuclear RNA genes U6 and H1 are commonly
used, as these promoters are ubiquitously expressed, resulting in high
level shRNA expression in most cell types, and have well-defined transcriptional start sites and stop signals (four to six consecutive thymidines) that yield a 2-nucleotide, 3 overhang. The U6 and H1
promoters have been used successfully to express shRNA in vivo (26,
5155) and the first one will be used in the protocol described here.
Selection of the hairpin sequence is the most important aspect
to prevent interferon responses. As mentioned above, there are
motifs that are known to stimulate interferon production (1215)
and these motifs should be avoided in the construct. Nevertheless,
even if these are avoided, it is still possible that an interferon
response will be induced from unknown motifs. Most constructs
are designed to target a 1921-nucleotide sequence of the mRNA.
A couple of approaches can be used to express shRNA. The first
one consists of generating a structure that has a stem with complete
homology to the target sequence (17, 47). This is the more
straightforward and the most commonly used approach. The second
involves a longer structure, with the features of a microRNA (45, 56,
57). The target sequence is embedded within the microRNA, and
after processing, the final siRNA is the same as a processed naked
shRNA. We describe the design of naked shRNA constructs.
1. The target transcript sequence can be obtained from the Entrez
database provided by the National Center for Biotechnology
Information (http://www.ncbi.nlm.nih.gov/Entrez). Use the
open reading frame (ORF) to select for potential sequences to
be targeted. If the ORF is short, the 5 and 3 untranslated
regions can also be included. For the beginner, several companies
(Ambion, Dharmacon, Invitrogen) offer shRNA libraries and
provide premade constructs. In the majority of cases, these
constructs have not been validated, though, and the investigator
is expected to test them. Alternatively, one can design the
hairpin using algorithms currently available on the websites
168
of the same companies. To use the BLOCK-iT U6 RNAi

Entry Vector Kit, go to http://rnaidesigner.invitrogen.com/
rnaiexpress/setOption.do?designOption=shrna&pid=2095707
992663696800 (BLOCK-iT RNAi Designer). Introduce the
nucleotide sequence or accession number and the species, and
select pENTR/U6 and sense/loop/antisense (choosing antisense/loop/sense will work equally well). The system includes a
Basic Local Alignment Search Tool (BLAST) search for potential
homologies to mRNAs other than the target. Click on RNAi
design. Once target sequences are obtained, choose the top
four on the list. A web-based algorithm to identify mRNAs with
seed matches (nucleotides 28 starting the 5 of the guide
strand) to the mature shRNA sequence can also be used to discard sequences that can potentially induce off-target effects
(http://www.dharmacon.com/seedlocator/seedlocatortemplate.
aspx). Sequences with the least number of matching nucleotides
in genes are potentially less susceptible to give rise to these effects.
Once four appropriate target sequences have been selected, click
Design shRNA oligos. Choose one of the default loop
sequences, and select Design. The sequence of the top and
bottom sequences will appear. Oligonucleotides should be
ordered at the 50 nmol scale, desalted, and unmodified.
2. Annealing and cloning into the plasmid is done following
the kit instructions. It is a straightforward sticky-ends
cloning step.
3. Transform E. coli subcloning grade bacteria, following the manufacturers recommendations. Use 24 mL of the ligation product and 50 mL of bacteria solution per construct. Include pUC18
as control for efficacy of the transformation, and a negative control without plasmid. Incubate the plates overnight at 37C.
4. For each construct, test four to five colonies. Grow each
colony in 3 mL LB broth containing 100 mg/mL ampicillin
for 16 h at 37C.
5. Prepare minipreps using the QIAprep Spin Miniprep Kit. The
DNA should be sequenced to determine the correct insert that
has been incorporated.
6. Testing the constructs for silencing efficacy can be done by
transfection of a cell line known to express the target gene. An
important point to keep in mind is that the level of transfection
should be at least 50%, otherwise it will be difficult to observe
silencing. Alternatively, a construct expressing the target gene
can be generated and co-transfection experiments in HEK293
cells can be designed. We have successfully knock down genes
following this approach. Plate 4 105 HEK293 cells in six-well
plates, using DMEM medium, supplemented with 10% (v/v)
FBS. The next day transfect cells using 0.5 mg plasmid expressing the target gene plus 1 mg plasmid expressing the shRNA.
10
169
Metafectene Pro is used at a 4:1 ratio. As control, use the plasmid

expressing a scrambled sequence. The level of knock-down is
compared to the scrambled construct.
7. Simultaneously to testing for efficacy to silence the target gene,
constructs can be tested for interferon response induction in
the same cell line (28, 33). Nevertheless, one has to keep in
mind that even if a negative result is observed, there is a possibility that the shRNA triggers interferon production in vivo,
as the response can differ among cell types, means of delivery
of the shRNA, and timing (58). In particular, when targeting
tissues that are rich in dendritic cells, macrophages, or other
immune cells, there is a possibility for activating RNA-binding
receptors present in these cells.
3.2. Viral Vector
Selection
For delivery of the shRNA expression cassette in vivo, a viral vector

is typically used. Several viral systems have been reported for the
use of shRNA, including retroviral and lentiviral vectors, adenovirus
vectors, and adeno-associated virus (AAV) vectors.
Retroviral and lentiviral vectors are useful tools for transducing
cells in studies that require integration of the shRNA-expression
cassette in the host genome (59, 60). Lentiviruses, in particular,
are optimal delivery vectors because they can transduce quiescent
as well as dividing cells. This vector system has been used for in vivo
(6164) and ex vivo (55, 65) approaches. Lentiviral vector systems
are available from Dharmacon, GenScript, and Invitrogen.
Adenovirus vectors can transduce a large variety of cell types
and have the advantage over other viral systems of being produced
at high titers. A few in vivo studies have shown it is possible to
obtain a significant level of gene silencing using first generation
(E1-deleted) vectors (22, 6668). However, a limitation of this
type of vector is that it is only suitable for short-term expression
and it expresses VA1 noncoding RNA, known to interfere with the
RNAi processing pathway (69). The latest generation of adenoviral
vectors has all viral genes deleted, leaving only the ITRs and packaging sequence [known as helper-dependent (HDAd), high-capacity or gutless adenovirus]. This more advanced vector system
yields long-term transgene expression in liver upon intravenous
administration (7072). Localized applications in the brain (7375),
muscle (76, 77), and lung (78) also resulted in persistent expression. We have used this system in liver to silence genes for several
weeks (26, 27). Adenoviral vectors can be generated using commercially available kits from Invitrogen, GenScript, Agilent
Technologies, and Microbix Biosystems.
AAVs are single-stranded DNA viruses highly used in the field
of gene therapy (79). AAV2 transduces a wide range of cell types,
and several other serotypes have been shown to transduce specific
tissues with greater efficiency. As described above for adenoviral
vectors, local injection of AAV in tissues results in long-term expression.
170
Disadvantages of this vector in the context of shRNA expression

include a significant delay in transgene expression on the order of
weeks depending on serotype (80). Also, AAV is known to form
high molecular weight DNA concatemers (81). This property
could make it difficult to determine optimal vector doses, and
result in toxicity as extremely high amounts of shRNA are expressed.
AAV shRNA-expressing vectors have been used in a number of
in vivo studies (53, 82, 83) and can be generated using commercially available kits from Cell Biolabs, Inc.
3.3. RNA Isolation
and qPCR
For accurate mRNA quantification of interferon-responsive genes,

forward and reverse primers are designed to generate an amplicon
of approximately 150250 bp and to bind different exons of the
gene. This prevents amplification from any potential DNA contaminating the RNA solution. Primer sequences for mouse OAS1b,
ISG56, and ISG20 are provided in Table 1, and are synthesized at
the 50 nmol scale. If setting up a new primer pair, use a primer
design application such as MacVector (Accelrys Inc., San Diego,
CA) to determine whether the sequences chosen are free from
hairpin structures, 3 self-dimers or duplex formation. As a normalizer, the b-actin, U6, or L32 ribosomal protein genes can be used.
We have found that b-actin and other cytoskeleton-related genes
are often upregulated in livers of animals that developed the IFN
response (probably as the result of increased cell death and subsequent hepatocyte repopulation), thereby impairing using these
genes as controls for RNA loading. The small nuclear gene U6 or
the L32 ribosomal proteins are alternative genes. However, a
Students t-test should be done to confirm that the gene used as
normalizer is not differentially expressed between groups. When
working with RNA, use gloves at all times and use solutions prepared with nuclease-free water to prevent RNA degradation.
1. Administer the viral vector in vivo, including the following groups:
vehicle, viral vector expressing a scrambled shRNA sequence
(shSCR), viral vector expressing the target shRNA sequence
(shTG), viral vector without an expression cassette (NEC).
2. The timing for sacrificing the animals is an important aspect.
Consider looking at multiple time points. The interferon
response is short-lived and absence of positive data does not
mean absence of the response at other time points. Using adenoviral vectors expressing shRNA in liver, we have detected
interferon responses 1 week after vector administration.
3. Total RNA is isolated from approximately 100 mg tissue. Treat
with DNase to remove contaminating DNA.
4. Using total RNA from a vehicle-treated sample, prepare a standard curve with 100, 50, 25, 12.5, and 6.25 ng/mL, by doing
serial 1:2 dilutions (e.g., transfer 20 mL of first dilution to a
clean tube with 20 mL of water, mix well, and discard tip.
AGAGATCACGGACTACAGAA
AGAGAACAGCTACCACCTTT
ISG20
ISG56
TGGACCTGCTCTGAGATTCT
TCTGTGGACGTGTCATAGAT
TGAGGCGCTTCAGCTTGGTT
TTGATGTGCTGCCAGCCTAT
OAS1b
ATGGCTGGGGTGTTGAAGGTC
ATCCTCTTGCCCTGATCCTT
CTACAATGAGCTGCGTGTGGC
b-Actin
Reverse primer (53)
L32 ribosomal ACATTTGCCCTGAATGTGGT

protein
Forward primer (53)
Gene
Table 1
Primer sequences
56
52
56
54
62
Annealing (C)
75
78
77
72
78
160
197
186
199
125
Data
Product
collection (C) size (bp)
(26)
(26)
(26)
(84)
(26)
References
10
171
172
Table 2
Reaction setup
Reagent
Volume
2 QuantiTect SYBR Green RT-PCR master mix
25 mL
Primer forward (25 pmol/mL)
1 mL
Primer reverse (25 pmol/mL)
1 mL
QuantiTect RT mix
0.5 mL
Template RNA (25 ng/mL)
2 mL
Nuclease-free water
20.5 mL
Final volume
50 mL
Repeat four times. Generating accurate standards is critical, as

the quality of the standard curve determines the success of the
assay). Prior to testing the samples, run the standards as
described below and confirm that only one fragment is obtained
by running the PCR product in a 1% agarose gel in 1 TBE.
5. Samples are diluted to 25 ng/mL and analysis is done using
2 mL of this solution.
6. Working solutions of primers are made by dilution to
25 pmol/mL in nuclease-free water. Use 0.5 mM of each primer
in a 50 mL reaction.
7. Standards, samples, and blank (nuclease-free water) are tested
in duplicate or triplicate (see Table 2). Prepare a second set of
samples without RT mix, to confirm amplification is from
RNA, instead of contaminating DNA. Program the ABI 7500
real-time PCR machine as follows: 30 min 50C (RT reaction);
15 min 95C (HotStarTaq DNA polymerase activation); 40
cycles of: 15 s 94C (denaturation), 30 s 5060C (annealing),
30 s 72C (extension), 15 s 7578C (data collection); add a
melting curve analysis step. If using a different instrument,
adjust these conditions as recommended by the manufacturer.
8. Results are expressed as amount of RNA relative to that present in
the sample used as standard. For normalization, divide the value of
the ISG gene by the value obtained for the gene used as normalizer. All groups are tested for statistical significance using a t-test
and the vehicle-treated group as reference. The virus not expressing an shRNA is a good control for the impact of the virus capsid
and genome on the innate immune response. Optimally, this group
should not display differences with the vehicle-treated group. If it
does, interpretation of the data becomes more complicated, as it is
not possible to discern between interferon induction from the
virus itself or the shRNA. The presence of an interferon response
10
173
Fig. 1. Interferon-stimulated gene (ISG) expression. (a) Mice were administered 1 1011 viral
particles of helper-dependent adenoviral vectors expressing shRNA against Sterol Regulatory
Element Binding Protein 1 (gAd.shSREBP1), expressing a scrambled sequence (gAd.shSCR), or
without expression cassette (gAd.NEC). A group of mice received vehicle. Mice were sacrificed
1, 3, and 6 weeks later. No activation of the interferon response was observed. (b) RNA was
isolated from livers of mice that received the doses shown at the bottom, and sacrificed 1
week later. Mice that received the gAd.sh242 vector (expressing an shRNA against fatty acid
binding protein 5, Fabp5), developed an interferon response. Reproduced with permission from
The Journal of Biological Chemistry (26). Values indicate the -fold level of expression relative to
the vehicle-treated group. Asterisk p < 0.05; n = 3.
174
can be detected as a fold increase in gene expression compared

to vehicle and NEC-treated group (Fig. 1).
9. In addition, tissue sections can be evaluated histologically to
look for signs of inflammation and cell death. We have found a
100% correlation between presence of a strong interferon
response and inflammatory foci in the liver of mice (26).
10. The presence of an interferon response can be further confirmed by analysis of liver function markers, such as ALT, from
serum. If studying a different tissue and reliable markers of cellular damage are available, those should be included in the
analysis as well.
Acknowledgments
This research was supported by grants from the NIDDK
(DK069432-01 and DK078595), American Diabetes Foundation
(1-08-RA-135), and INGEN (Indiana Genomics Initiative of
Indiana University supported in part by Lilly Endowment Inc.).
References
1. Sledz, C.A., Williams, B.R. (2004) RNA interference and double-stranded-RNA-activated
pathways. Biochem. Soc. Trans. 32: 952956.
2. Sledz, C.A., Williams, B.R. (2005) RNA interference in biology and disease. Blood 106:
787794.
3. Stark, G.R., Kerr, I.M., Williams, B.R.,
Silverman, R.H., Schreiber, R.D. (1998) How
cells respond to interferons. Annu. Rev.
Biochem. 67: 227264.
4. Li, G., Xiang, Y., Sabapathy, K., Silverman,
R.H. (2004) An apoptotic signaling pathway in
the interferon antiviral response mediated by
RNase L and c-Jun NH2-terminal kinase.
J. Biol. Chem. 279: 11231131.
5. Alexopoulou, L., Holt, A.C., Medzhitov, R.,
Flavell, R.A. (2001) Recognition of doublestranded RNA and activation of NF-kappaB by
Toll-like receptor 3. Nature 413: 732738.
6. Diebold, S.S., Kaisho, T., Hemmi, H., Akira,
S., Reis e Sousa, C. (2004) Innate antiviral
responses by means of TLR7-mediated recognition of single-stranded RNA. Science 303:
15291531.
7. Kariko, K., Bhuyan, P., Capodici, J., Weissman,
D. (2004) Small interfering RNAs mediate
sequence-independent gene suppression and
induce immune activation by signaling through
toll-like receptor 3. J. Immunol. 172:
65456549.
8. Melchjorsen, J., Jensen, S.B., Malmgaard, L.,

et al. (2005) Activation of innate defense
against a paramyxovirus is mediated by RIG-I
and TLR7 and TLR8 in a cell-type-specific
manner. J. Virol. 79: 1294412951.
9. Barber, G.N. (2005) The dsRNA-dependent
protein kinase, PKR and cell death. Cell Death
Differ. 12: 563570.
10. Yoneyama, M., Kikuchi, M., Matsumoto, K.,
et al. (2005) Shared and unique functions of
the DExD/H-box helicases RIG-I, MDA5,
and LGP2 in antiviral innate immunity.
J. Immunol. 175: 28512858.
11. Kato, H., Takeuchi, O., Sato, S., et al. (2006)
Differential roles of MDA5 and RIG-I helicases
in the recognition of RNA viruses. Nature
441: 101105.
12. Hornung, V., Guenthner-Biller, M., Bourquin,
C., et al. (2005) Sequence-specific potent
induction of IFN-alpha by short interfering
RNA in plasmacytoid dendritic cells through
TLR7. Nat. Med. 11: 263270.
13. Judge, A.D., Sood, V., Shaw, J.R., Fang, D.,
McClintock, K., MacLachlan, I. (2005)
Sequence-dependent stimulation of the mammalian innate immune response by synthetic
siRNA. Nat. Biotechnol. 23: 457462.
14. Forsbach, A., Nemorin, J.G., Volp, K., et al.
(2007) Characterization of conserved viral
leader RNA sequences that stimulate innate
10
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
immunity through TLRs. Oligonucleotides 17:

405417.
Forsbach, A., Nemorin, J.G., Montino, C.,
et al. (2008) Identification of RNA sequence
motifs stimulating sequence-specific TLR8dependent immune responses. J. Immunol.
180: 37293738.
Elbashir, S.M., Harborth, J., Lendeckel, W.,
Duplexes of 21-nucleotide RNAs mediate RNA
interference in cultured mammalian cells.
Nature 411: 494498.
Brummelkamp, T.R., Bernards, R., Agami, R.
(2002) A system for stable expression of short
interfering RNAs in mammalian cells. Science
296: 550553.
Sui, G., Soohoo, C., Affar el, B., Gay, F., Shi,
Y., Forrester, W.C. (2002) A DNA vectorbased RNAi technology to suppress gene
expression in mammalian cells. Proc. Natl.
Acad. Sci. USA 99: 55155520.
Arts, G.J., Langemeijer, E., Tissingh, R., et al.
(2003) Adenoviral vectors expressing siRNAs
for discovery and validation of gene function.
Genome Res. 13: 23252332.
Lewis, D.L., Hagstrom, J.E., Loomis, A.G.,
Wolff, J.A., Herweijer, H. (2002) Efficient
delivery of siRNA for inhibition of gene expression in postnatal mice. Nature Genet. 32:
107108.
McCaffrey, A.P., Meuse, L., Pham, T.T.,
Conklin, D.S., Hannon, G.J., Kay, M.A. (2002)
RNA interference in adult mice. Nature 418:
3839.
Xia, H., Mao, Q., Paulson, H.L., Davidson,
B.L. (2002) siRNA-mediated gene silencing
in vitro and in vivo. Nat. Biotechnol. 20:
10061010.
Hommel, J.D., Sears, R.M., Georgescu, D.,
Simmons, D.L., DiLeone, R.J. (2003) Local gene
knockdown in the brain using viral-mediated
RNA interference. Nat. Med. 9: 15391544.
Kishida, T., Asada, H., Gojo, S., et al. (2004)
Sequence-specific gene silencing in murine
muscle induced by electroporation-mediated
transfer of short interfering RNA. J. Gene. Med.
6: 105110.
Matsuda,
T.,
Cepko,
C.L.
(2004)
Electroporation and RNA interference in the
rodent retina in vivo and in vitro. Proc. Natl.
Acad. Sci. USA 101: 1622.
Witting, S.R., Brown, M., Saxena, R., Nabinger,
S., Morral, N. (2008) Helper-dependent
Adenovirus-mediated Short Hairpin RNA
Expression in the Liver Activates the Interferon
Response. J. Biol. Chem. 283: 21202128.
175
27. Ruiz, R., Witting, S.R., Saxena, R., Morral, N.

(2009) Robust hepatic gene silencing for functional studies using helper-dependent adenovirus vectors. Hum. Gene Ther. 20: 8794.
28. Bridge, A.J., Pebernard, S., Ducraux, A.,
Nicoulaz, A.L., Iggo, R. (2003) Induction of
an interferon response by RNAi vectors in mammalian cells. Nature Genet. 34: 263264.
29. Sledz, C.A., Holko, M., de Veer, M.J.,
Silverman, R.H., Williams, B.R. (2003)
Activation of the interferon system by shortinterfering RNAs. Nat. Cell Biol. 5: 834839.
30. Fish, R.J., Kruithof, E.K. (2004) Short-term
cytotoxic effects and long-term instability of
RNAi delivered using lentiviral vectors. BMC
Mol. Biol. 5: 9.
31. Pebernard, S., Iggo, R.D. (2004) Determinants
of interferon-stimulated gene induction by
RNAi vectors. Differentiation 72: 103111.
32. Gantier, M.P., Williams, B.R. (2007) The
response of mammalian cells to double-stranded
RNA. Cytokine Growth Factor Rev. 18:
363371.
33. Kenworthy, R., Lambert, D., Yang, F., et al.
(2009) Short-hairpin RNAs delivered by lentiviral vector transduction trigger RIG-Imediated IFN activation. Nucleic Acids Res.
37: 65876599.
34. Judge, A.D., Bola, G., Lee, A.C., MacLachlan,
I. (2006) Design of noninflammatory synthetic
siRNA mediating potent gene silencing in vivo.
Mol. Ther. 13: 494505.
35. Hornung, V., Ellegast, J., Kim, S., et al. (2006)
5-Triphosphate RNA is the ligand for RIG-I.
Science 314: 994997.
36. Marques, J.T., Devosse, T., Wang, D., et al.
(2006) A structural basis for discriminating
between self and nonself double-stranded
RNAs in mammalian cells. Nat. Biotechnol. 24:
559565.
37. Pichlmair, A., Schulz, O., Tan, C.P., et al.
(2006) RIG-I-mediated antiviral responses to
single-stranded RNA bearing 5-phosphates.
Science 314: 9971001.
38. Malarkey, D.E., Johnson, K., Ryan, L.,
Boorman, G., Maronpot, R.R. (2005) New
insights into functional aspects of liver morphology. Toxicol. Pathol. 33: 2734.
39. Stratford-Perricaudet, L.D., Levrero, M.,
Chase, J.F., Perricaudet, M., Briand, P. (1990)
Evaluation of the transfer and expression in
mice of an enzyme-encoding gene using a
human adenovirus vector. Hum. Gene Ther. 1:
241256.
40. Morral, N., ONeal, W.K., Rice, K., et al.
(2002) Lethal toxicity, severe endothelial
176
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
51.

injury, and a threshold effect with high doses of
an adenoviral vector in baboons. Hum. Gene
Ther. 13: 143154.
Inagaki, Y., Kushida, M., Higashi, K., et al.
(2005) Cell type-specific intervention of transforming growth factor beta/Smad signaling
suppresses collagen gene expression and hepatic
fibrosis in mice. Gastroenterology 129:
259268.
Wheeler, M.D., Yamashina, S., Froh, M.,
Rusyn, I., Thurman, R.G. (2001) Adenoviral
gene delivery can inactivate Kupffer cells: role
of oxidants in NF-kappaB activation and
cytokine production. J. Leukoc. Biol. 69:
622630.
Manickan, E., Smith, J.S., Tian, J., et al. (2006)
Rapid Kupffer cell death after intravenous
injection of adenovirus vectors. Mol. Ther. 13:
108117.
Nguyen, L.H., Espert, L., Mechti, N., Wilson,
D.M., 3 rd. (2001) The human interferon- and
estrogen-regulated ISG20/HEM45 gene
product degrades single-stranded RNA and
DNA in vitro. Biochemistry 40: 71747179.
Zeng, Y., Wagner, E.J., Cullen, B.R. (2002)
Both natural and designed micro RNAs can
inhibit the expression of cognate mRNAs when
expressed in human cells. Mol. Cell 9:
13271333.
Denti, M.A., Rosa, A., Sthandier, O., De
Angelis, F.G., Bozzoni, I. (2004) A new vector, based on the PolII promoter of the U1
snRNA gene, for the expression of siRNAs in
mammalian cells. Mol. Ther. 10: 191199.
Paddison, P.J., Caudy, A.A., Bernstein, E.,
Hannon, G.J., Conklin, D.S. (2002) Short
hairpin RNAs (shRNAs) induce sequencespecific silencing in mammalian cells. Genes
Dev. 16: 948958.
Stegmeier, F., Hu, G., Rickles, R.J., Hannon,
G.J., Elledge, S.J. (2005) A lentiviral
microRNA-based system for single-copy polymerase II-regulated RNA interference in mammalian cells. Proc. Natl. Acad. Sci. USA 102:
1321213217.
Zeng, Y., Yi, R., Cullen, B.R. (2005)
Recognition and cleavage of primary microRNA
precursors by the nuclear processing enzyme
Drosha. Embo J. 24: 138148.
Zeng, Y., Cai, X., Cullen, B.R. (2005) Use of
RNA polymerase II to transcribe artificial
microRNAs. Methods Enzymol. 392: 371380.
Dentin, R., Benhamed, F., Hainault, I., et al.
(2006) Liver-specific inhibition of ChREBP
improves hepatic steatosis and insulin resistance
in ob/ob mice. Diabetes 55: 21592170.
52. Xu, H., Wilcox, D., Nguyen, P., et al. (2006)

Hepatic knockdown of mitochondrial GPAT1
in ob/ob mice improves metabolic profile.
Biochem. Biophys. Res. Commun. 349:
439448.
53. Chen, C.C., Ko, T.M., Ma, H.I., et al. (2007)
Long-term inhibition of hepatitis B virus in
transgenic mice by double-stranded adenoassociated virus 8-delivered short hairpin RNA.
Gene Ther. 14: 1119.
54. Huang, B., Schiefer, J., Sass, C., Landwehrmeyer,
G.B., Kosinski, C.M., Kochanek, S. (2007)
High-capacity adenoviral vector-mediated
reduction of huntingtin aggregate load in vitro
and in vivo. Hum. Gene Ther. 18: 303311.
55. Ohmori, T., Kashiwakura, Y., Ishiwata, A.,
Madoiwa, S., Mimuro, J., Sakata, Y. (2007)
Silencing of a targeted protein in in vivo platelets using a lentiviral vector delivering short
hairpin RNA sequence. Arterioscler. Thromb.
Vasc. Biol. 27: 22662272.
56. McManus, M.T., Petersen, C.P., Haines, B.B.,
Chen, J., Sharp, P.A. (2002) Gene silencing
using micro-RNA designed hairpins. RNA 8:
842850.
57. Silva, J.M., Li, M.Z., Chang, K., et al. (2005)
Second-generation shRNA libraries covering
the mouse and human genomes. Nature Genet.
37: 12811288.
58. Robbins, M., Judge, A., MacLachlan, I. (2009)
siRNA and innate immunity. Oligonucleotides
19: 89102.
59. Purdy, A.K., Campbell, K.S. (2010)
Introduction of shRNAs into human NK-like
cell lines with retrovirus. Methods Mol. Biol.
612: 223231.
60. Liu, Y.P., von Eije, K.J., Schopman, N.C., et al.
(2009) Combinatorial RNAi against HIV-1
using extended short hairpin RNAs. Mol. Ther.
17: 17121723.
61. Georgievska, B., Jakobsson, J., Persson, E.,
Ericson, C., Kirik, D., Lundberg, C. (2004)
Regulated delivery of glial cell line-derived
neurotrophic factor into rat striatum, using a
tetracycline-dependent lentiviral vector. Hum.
Gene Ther. 15: 934944.
62. Makinen, P.I., Koponen, J.K., Karkkainen,
A.M., et al. (2006) Stable RNA interference:
comparison of U6 and H1 promoters in
endothelial cells and in mouse brain. J. Gene
Med. 8: 433441.
63. Chen, Y., Lin, M.C., Yao, H., et al. (2007)
Lentivirus-mediated RNA interference targeting
enhancer of zeste homolog 2 inhibits hepatocellular carcinoma growth through down-regulation
of stathmin. Hepatology 46: 200208.
10
64. Couturier, C., Sarkis, C., Seron, K., et al.

(2007) Silencing of OB-RGRP in mouse
hypothalamic arcuate nucleus increases leptin
receptor signaling and prevents diet-induced
obesity. Proc. Natl. Acad. Sci. USA 104:
1947619481.
65. Bot, I., Guo, J., Van Eck, M., et al. (2005)
Lentiviral shRNA silencing of murine bone
marrow cell CCR2 leads to persistent knockdown of CCR2 function in vivo. Blood 106:
11471153.
66. Taniguchi, C.M., Ueki, K., Kahn, R. (2005)
Complementary roles of IRS-1 and IRS-2 in
the hepatic regulation of metabolism. J. Clin.
Invest. 115: 718727.
67. Narvaiza, I., Aparicio, O., Vera, M., et al.
(2006) Effect of adenovirus-mediated RNA
interference on endogenous microRNAs in a
mouse model of multidrug resistance protein 2
gene silencing. J. Virol. 80: 1223612247.
68. Sakamoto, N., Tanabe, Y., Yokota, T., et al.
(2008) Inhibition of hepatitis C virus infection
and expression in vitro and in vivo by recombinant adenovirus expressing short hairpin RNA.
J. Gastroenterol. Hepatol. 23: 14371447.
69. Lu, S., Cullen, B.R.. (2004) Adenovirus VA1
noncoding RNA can inhibit small interfering
RNA and MicroRNA biogenesis. J. Virol. 78:
1286812876.
70. Morral, N., Parks, R., Zhou, H., et al. (1998)
High doses of a helper-dependent adenoviral
vector yield supraphysiological levels of a1-antitrypsin with negligible toxicity. Hum. Gene
Ther. 9: 27092716.
71. Morsy, M.A., Gu, M.C., Motzel, S., et al.
(1998) An adenoviral vector deleted for all viral
coding sequences results in enhanced safety and
extended expression of a leptin transgene. Proc.
72. Schiedner, G., Morral, N., Parks, R., et al.
(1998) Genomic DNA transfer with a highcapacity adenovirus vector results in improved
in vivo gene expression and decreased toxicity.
Nature Genet. 18: 180183.
73. Thomas, C.E;, Schiedner, G., Kochanek, S.,
Castro, M.G., Lowenstein, P.R. (2000) Peripheral
infection with adenovirus causes unexpected
long-term brain inflammation in animals
injected intracranially with first-generation, but
not with high-capacity, adenovirus vectors:
toward realistic long-term neurological gene
74.
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
177
therapy for chronic diseases. Proc. Natl. Acad.

Sci. USA 97: 74827487.
Zou, L., Yuan, X., Zhou, H., Lu, H., Yang, K.
(2001) Helper-dependent adenoviral vectormediated gene transfer in aged rat brain. Hum.
Gene Ther. 12: 181191.
Xiong, W., Goverdhana, S., Sciascia, S.A., et al.
(2006) Regulatable gutless adenovirus vectors
sustain inducible transgene expression in the
brain in the presence of an immune response
against adenoviruses. J. Virol. 80: 2737.
Clemens, P.R., Kochanek, S., Sunada, Y., et al.
(1996) In vivo muscle gene transfer of fulllength dystrophin with an adenoviral vector
that lacks all viral genes. Gene Ther. 3:
965972.
Dudley, R.W., Lu, Y., Gilbert, R., et al. (2004)
Sustained improvement of muscle function one
year after full-length dystrophin gene transfer into
mdx mice by a gutted helper-dependent adenoviral vector. Hum. Gene Ther. 15: 145156.
Toietta, G., Koehler, D.R., Finegold, M.J.,
Lee, B., Hu, J., Beaudet, A.L. (2003) Reduced
inflammation and improved airway expression
using helper-dependent adenoviral vectors with
a K18 promoter. Mol. Ther. 7: 649658.
Carter, B.J. (2005) Adeno-associated virus vectors in clinical trials. Hum. Gene Ther. 16:
541550.
Wu, Z., Asokan, A., Samulski, R.J. (2006)
Adeno-associated virus serotypes: vector toolkit for human gene therapy. Mol. Ther. 14:
316327.
Carter, B.J. (2003) Gene delivery technology:
adeno-associated virus. In Pharmaceutical gene
delivery systems (S.M. Rolland AaS Ed.),
Marcel Dekker, Inc., New York pp. 183213.
Wang, X., Skelley, L., Cade, R., Sun, Z. (2006)
AAV delivery of mineralocorticoid receptor
shRNA prevents progression of cold-induced
hypertension and attenuates renal damage.
Gene Ther. 13: 10971103.
Paskowitz, D.M., Greenberg, K.P., Yasumura,
D., et al. (2007) Rapid and stable knockdown
of an endogenous gene in retinal pigment epithelium. Hum. Gene Ther. 18: 871880.
Zhang, W., Patil, S., Chauhan, B., et al. (2006)
FoxO1 regulates multiple metabolic pathways
in the liver: effects on gluconeogenic, glycolytic, and lipogenic gene expression. J. Biol.
Chem. 281: 1010510117.
Chapter 11
Use of RNA Interference to Investigate Cytokine Signal
Transduction in Pancreatic Beta Cells
Fabrice Moore, Daniel A. Cunha, Hindrik Mulder, and Decio L. Eizirik
Abstract
Type 1 diabetes (T1D) is a chronic autoimmune disease characterized by immune infiltration of the
pancreatic islets resulting in an inflammatory reaction named insulitis and subsequent beta cell apoptosis.
During the course of insulitis beta cell death is probably caused by direct contact with activated macrophages and T-cells, and/or exposure to soluble mediators secreted by these cells, including cytokines,
nitric oxide, and free oxygen radicals. In vitro exposure of beta cells to the cytokines interleukin(IL)1 + interferon(IFN)- or to tumor necrosis factor(TNF)- + IFN- induces beta cell dysfunction and ultimately apoptosis. The transcription factors NF-B and STAT1 are key regulators of cytokine-induced beta
cell death. However, little is known about the gene networks regulated by these (or other) transcription
factors that trigger beta cell apoptosis. The recent development of RNA interference (RNAi) technology
offers a unique opportunity to decipher the cytokine-activated molecular pathways responsible for beta cell
death. Use of RNAi has been hampered by technical difficulties in transfecting primary beta cells, but in
recent years we have succeeded in developing reliable and reproducible protocols for RNAi in beta cells.
This chapter details the methods and settings used to achieve efficient and nontoxic transfection of small
interfering RNA in immortal and primary beta cells.
Key words: Small interfering RNA, siRNA, Pancreatic beta cells, Apoptosis, Gene knockdown,
Inducible nitric oxide synthase, Interleukin-1, Interferon-, Tumor necrosis factor-
1. Introduction
Research on RNA interference (RNAi) gained momentum in
1998, when Fire and Mello reported that double-stranded (ds)
RNA induced gene silencing in Caenorhabditis elegans (1). Since
then, there has been a growing understanding of the complexity of
RNA-based gene silencing, including the discovery of several subclasses of small interfering (si)RNAs in plants, fungi, and mammals
179
180
F. Moore et al.
(reviewed in ref. 2, 3). SiRNA-based gene knockdown is mediated

via the enzymatic generation of short RNA sequences from a larger
dsRNA precursor that specifically recognize and degrade sequencematched target mRNAs (2). Typically, dsRNAs are processed by
the RNAse Dicer (or sometimes by another RNAse) into 2124
nucleotide duplexes with two nucleotides overhanging at 3ends.
The guide strand of the siRNA duplex is then loaded into a multiprotein complex, or RNA-induced silencing complex (RISC),
which identifies the complementary mRNA to be targeted for degradation, ultimately resulting in the decreased expression of the
target protein (reviewed in ref. 2, 3). The precursor dsRNA may be
endogenously generated, but it can also be derived from exogenous sources; this opened novel possibilities for the experimental
modulation of gene expression. Indeed, synthetic siRNAs have
been shown to potently inhibit gene expression in mammalian cells
(4), and recent advances in systemic and/or local siRNA delivery
opened new perspectives for their future use in human therapy (5, 6).
Furthermore, synthetic siRNAs have proved very useful under
in vitro conditions to decipher the role(s) of different genes in
diverse cellular processes (79).
Our laboratory has been studying for many years the gene networks and molecular pathways involved in cytokine-induced pancreatic beta cell apoptosis (10), but these studies were often
hampered by the lack of fast and efficient methods to knockdown
gene and protein expression in beta cells. Indeed, beta cells are
hard-to-transfect cells, and do not tolerate well infection by adenoviral vectors at MOIs above 1020. To obviate this problem, we
have tested different approaches to efficiently introduce synthetic
siRNAs in rat and human beta cells. We describe below the materials, reagents, and methods presently used in our laboratory to
achieve efficient and nontoxic siRNA-mediated gene knockdown
in both primary rat and human beta cells, and in insulin-producing
cell lines.
2. Materials
2.1. Synthetic siRNAs
and Transfection
Reagents
1. siGLO Green Transfection Indicator (referred to as FITCconjugated siRNA Thermo Scientific, Lafayette, CO, USA),
used to assess the efficiency of transfection, is reconstituted
with double distilled sterile water (ddH2O) at a concentration
of 20 M, aliquoted, and snap frozen at 80C (see Note 1).
2. Allstars Negative Control siRNA (siCtrl Qiagen, Venlo, The
Netherlands) is reconstituted with the provided siRNA dilution buffer at 20 M, aliquoted, and stored at 80C.
11 Use of RNA Interference to Investigate Cytokine Signal Transduction...
181
3. HP GenomeWide siRNA rat inducible nitric oxide synthase

(referred to as si iNOS Qiagen) is reconstituted at 20 M in
siRNA dilution buffer, aliquoted, and stored at 80C.
4. The lipid-carrier DharmaFECT 1 Transfection Reagent was
purchased from Thermo Scientific and stored at 4C. This
transfection reagent was selected as the best one after testing
several other methods for siRNA transfection, including several other lipid reagents and electroporation.
5. OPTI-MEM medium (GIBCO, Bethesda, MD, USA) is used
without additives (see Note 2).
2.2. Cell Culture
and Transfection
1. PBS 1: 137 mM NaCl, 2.68 mM KCl, 4.29 mM Na2HPO4,

1.47 mM KH2PO4, pH 7.4.
2. Solution of trypsin at 0.5 mg/mL (from Cambrex, East
Rutherford, NJ, USA).
3. The rat insulin-producing INS-1E cell line (a kind gift from
Dr. C. Wollheim, Centre Medical Universitaire, Geneva,
Switzerland) is cultured in RPMI 1640 GlutaMAX-I, 5% (v/v)
fetal bovine serum, 10 mM HEPES, 1 mM Na-pyruvate,
50 M 2-mercaptoethanol (all from GIBCO), and 100 U/mL
penicillin + 100 g/mL streptomycin (Cambrex). For transfection, the same medium is used but without antibiotics (see
Note 3).
4. Male Wistar rats (Charles River Laboratories, Brussels,
Belgium) are housed and used according to the guidelines of
the Belgian Regulations for Animal Care. Islets are isolated by
collagenase digestion and hand picked under a stereomicroscope (11, 12).
5. Rat pancreatic islets are washed with Solution I (124 mM
NaCl, 5.4 mM KCl, 0.8 mM H2SO4, 1 mM NaH2PO4,
0.71 mM NaHCO3, 5 mM glucose, and 1 mM EGTA), and
then dispersed by incubation for 12 min in a solution of dispase (5 U/mL in Solution I, Roche Diagnostics, Vilvoorde,
Belgium). Beta cells are purified by autofluorescence-activated
cell sorting (FACSAria, BD Bioscience., San Jose, CA, USA)
(11, 12). Purified cells are cultured attached to polylysinecoated 96-well plates (BD Falcon, Franklin Lakes, NJ, USA) in
Hams F-10 medium containing 10 mM glucose, 50 M
3-isobutyl-1-methylxanthine (both from Sigma, Bornem,
Belgium), 2 mM glutaMAX, 5% (v/v) FBS (both from
GIBCO), 0.5% charcoal-absorbed BSA (Fraction V; Boehringer,
Indianapolis, IN, USA), 50 U/mL penicillin, and 50 g/mL
streptomycin (12, 13). For transfection, the same medium is
used but without antibiotics (see Note 3) or BSA (see Note 4).
Of note, in case FACS sorting facilities are not available, a similar
182
F. Moore et al.
transfection and culture procedure can be used with dispersed

islet cells.
6. Human islets are isolated as previously described (14) and dispersed by incubation for 23 min in a solution of dispase
(2.5 U/mL in Solution I Roche). Dispersed human islets are
cultured attached to polylysine-coated 96-well plates (BD
Falcon) in Hams F-10 containing 6.1 mM glucose, 10% (v/v)
FBS, 2 mM GlutaMAX, 50 M 3-isobutyl-1-methylxanthine,
1% (w/v) BSA, 50 U/mL penicillin, and 50 g/mL streptomycin (15). For transfection, the same medium described
above for rat beta cell transfection is used.
7. Recombinant human IL-1 (specific activity 1.8 107 U/mg;
a kind gift from C. W. Reinolds, National Cancer Institute,
Bethesda, MD, USA) is used at 100 U/mL.
2.3. Assessment
of Cell Viability
and Efficiency
of Transfection
1. Propidium iodide (PI, Sigma) is diluted in PBS at 1 mg/mL

for the stock solution, filtered through a 0.22 m filter, and
used at a final concentration of 5 g/mL.
2. The stock solution of Hoechst 33342 (HO, Sigma) is diluted
at 1 mg/mL in PBS, filtered, and used at a final concentration
of 5 g/mL.
3. siGLO Green Transfection indicator (see Subheading 2.1).
4. An inverted microscope (Zeiss, Zaventem, Belgium) with filters for excitation at 358 nm (HO), 538 nm (PI), and 495 nm
(FITC) is used.
2.4. Immunostaining
for Insulin
1. Sterile glass chamber slides and microscope coverslips (Thermo

Fisher Scientific, Rochester, NY, USA).
2. Formaldehyde solution: 4% buffered, pH 6.9, at room
temperature.
3. Washing buffer: PBS 1, pH 7.4 (see Subheading 2.2).
4. Permeabilization solution: 0.3% (v/v) Triton X-100 (Sigma)
in PBS.
5. Blocking solution: PBS with 10% (v/v) normal goat serum
(Gibco) and 0.8% (w/v) BSA (Fraction V; Boehringer).
6. Primary antibody: mouse monoclonal anti-insulin (1:1,000)
(Sigma).
7. Rhodamine-conjugated anti-mouse IgG (1:200) (Lucron
Bioproducts).
8. Hoechst 33342 solution (see Subheading 2.3).
9. Fluorescence mounting medium (Dakocytomation, Glostrup,
Denmark).
10. Microscope (see above).
183
3. Methods
3.1. Designing
Efficient siRNAs
and Selecting
the Adequate Control
Conditions
In recent years, major advances have been made in understanding

how to design efficient and specific siRNAs (reviewed in ref. 16,
17). Important points to keep in mind when designing siRNAs
include avoiding palindromic sequences and high GC-containing
duplexes, and favoring a poorly stable 5 antisense end of the siRNA
to ease the unwinding of the duplex and promote the loading of
the guide strand into the RISC complex (17). In order to ensure
the specificity of the gene knockdown (and thus decrease the risk
of off-target effects), it is recommended to both respect a perfect match between the siRNA and the targeted mRNA sequence,
and also to BLAST the siRNA sequence against the whole transcriptome to ensure that only one gene is targeted (17, 18). SiRNA
duplexes over 23 nucleotides have been shown to exhibit RNAiindependent off-target effects in some cell lines through activation
of the dsRNA recognition pathway (19). Various chemical modifications of the siRNA duplexes have been reported to improve their
stability and specificity (2022) and there are proprietary and nondisclosed modifications that are used by different industrial manufacturers. Many of these companies now offer siRNA designing
software available online. The algorithms used by such softwares
take into account the different criteria detailed above, and together
with the chemical modifications added to the siRNA duplexes allow
the design of highly effective and specific siRNAs (see Note 5).
Importantly, it is recommended to perform siRNA-based experiments using two (or more) siRNAs recognizing different regions
of the target mRNA sequence (23). This simple rule decreases the
risk of drawing inappropriate conclusions due to off-target effects
of a single siRNA. Moreover, siRNA experiments should include
both nontransfected cells and cells transfected with a nonbiologically active control siRNA. In early RNAi development, it was
assumed that the best control was a scramble sequence of the
original siRNA, including two or three nucleotide mismatches.
These scrambled controls, however, may pose serious problems,
since (1) the 23 nucleotide change in the middle of the siRNA
sequence may convert the siRNA into a miRNA that still inhibits
the translation of the targeted gene; (2) the changes in the original
siRNA sequence may generate an siRNA that inhibits other genes
in an unexpected way (23). Nowadays, many sequence-independent control siRNAs are commercially available. The sequences of
these control duplexes have been BLAST tested to minimize
sequence homology to any known gene in a given species. It is
generally recommended, however, that this control siRNA matches
the GC content of the original siRNA (see Note 6).
184
F. Moore et al.
3.2. Preparation
of the Cells
for the Transfection
3.2.1. INS-1E Cells
1. The rat insulinoma INS-1E cells are cultured in an incubator at

37C and 5% (v/v) CO2 until reaching ~80% confluence. The
cells are then washed once with PBS, detached with trypsin
and washed once with antibiotic-free INS-1E medium to
remove the trypsin (see Note 3).
2. After counting, INS-1E cells are plated at 104 cells/well in
a 96-well plate for viability experiments and 105 cells/well in a
24-well plate for RNA extraction or Western blot experiments.
3. The cells are cultured for a recovery period of 48 h in antibioticfree medium prior to transfection.
3.2.2. Rat Primary Beta

Cells and Human
Dispersed Islets
1. FACS-purified beta cells or dispersed human islets are obtained

as described in, respectively (11, 12) and (8), and plated at
2 104 cells/well in 96-well plates with the appropriate medium
(see Subheading 2.2).
2. After 24 h, the medium is removed and replaced by medium
without antibiotic and BSA, which is suitable for transfection
of both primary rat beta cells and dispersed human islets (see
Note 7).
3.3. Transfection
of siRNAs
This protocol is adapted for the transfection of 30 nM siRNA, a

concentration which we found in doseresponse experiments using
several siRNAs to provide potent RNAi without toxicity (Moore
et al., unpublished data). For certain targets, higher concentrations
(<100 nM) may be required. One may also consider using two different siRNAs to yield a greater efficiency. However, high concentrations of siRNA require careful control of toxicity and/or off
target effects.
3.3.1. INS-1E Cells
1. OPTI-MEM and antibiotic-free INS-1E medium are prewarmed at 37C.

2. siRNAs are thawed on ice; DharmaFECT lipid reagent is also
kept on ice.
3. The siRNAs is diluted at 300 nM (see Note 8; important) in
OPTI-MEM. From a 20 M stock siRNA, this means a 66.6
dilution (1 L of siRNA in 65.6 L of OPTI-MEM).
4. The DharmaFECT 1 lipid reagent is differentially diluted for
transfection in a 96- or a 24-well (see Note 9). For a 96-well,
0.1 L/well is used (1 L of DharmaFECT is diluted in 99 L
of OPTI-MEM), whereas 0.7 L/well is used for a 24-well
(1 L of DharmaFECT in 70.5 L of OPTI-MEM). The
diluted siRNA and DharmaFECT are kept separately for 5 min
at room temperature.
5. The diluted siRNA and DharmaFECT are mixed at a ratio of
1:1.
185
(a) For 96-well, 10 L of diluted siRNA is mixed with 10 L

of diluted DharmaFECT (20 L).
(b) For 24-well, 50 L of diluted siRNA is mixed with 50 L
of diluted DharmaFECT (100 L).
6. The mix is gently pipetted up and down three to four times
and incubated for 20 min at room temperature without
agitation.
7. The siRNA/lipid complexes are diluted 1:5 with antibioticfree INS-1E medium.
(a) For 96-well, 80 L of medium is added to the 20 L of
siRNA/lipid mix (100 L final).
(b) For 24-well, 400 L of medium is added to the 100 L of
siRNA/lipid mix (500 L final).
8. The culture medium is removed and the transfection medium
(100 L for a 96-well, and 500 L for a 24-well) is added for
overnight incubation (see Note 10).
9. The next day, the transfection medium is removed and replaced
by fresh antibiotic-free INS-1E medium for a recovery period
(see Note 11 and below).
3.3.2. Primary Rat Beta
Cells and Human
Dispersed Islets
The protocol used to transfect primary rat beta cells and human
dispersed islets is similar to the one used for transfection of INS-1E
cells in 96-wells, but with three changes:
1. Primary cells require a higher concentration of DharmaFECT
than INS-1E cells to be efficiently transfected (see Note 9).
(a) For primary rat beta cells, 0.25 L/well is used (2.5 L of
DharmaFECT is diluted in 97.5 L of OPTI-MEM).
(b) For dispersed human islet cells, 0.35 L/well is used
(3.5 L of DharmaFECT is diluted in 96.5 L of
OPTI-MEM).
2. The siRNA/lipid complexes are diluted 1:5 with the
antibiotic- and BSA-free transfection medium described in
Subheading 2.2.
3. After overnight transfection, the transfection medium is
removed and replaced by culture medium for respectively primary rat beta cells or human dispersed islet cells for recovery.
3.4. Posttransfection
Recovery
In RNAi experiments, it is important to allow an adequate posttransfection recovery for the following reasons: (a) the cells need
time to recover from the stressful process of transfection before
being challenged with another potential stressful treatment, such
as cytokine exposure; (b) an adequate recovery period is required
to achieve full inhibitory effects of the siRNA on the target gene/
protein. This is especially important when dealing with stable target
186
F. Moore et al.
mRNA/proteins. Under our experimental conditions, a recovery

period of 24 h post-siRNA transfection is usually sufficient to
observe a clear inhibition of de novo expressed proteins, for
instance to prevent iNOS expression following IL-1 treatment
(see below). On the other hand, and depending on the turnover
rate of the target protein, 4896 h of recovery should be allowed
to achieve inhibition of proteins that are expressed in the cells
under basal condition (see Note 12). For instance, 4896 h of
posttransfection period are required to observe a >75% inhibition in PTPN2, IKK, and MDA5 proteins (24).
3.5. Evaluation
of siRNA Transfection
Efficiency and Toxicity
A simple approach to evaluate siRNA transfection efficiency and

toxicity is to use as probe a fluorescence-labeled siRNA and to
evaluate cell viability with nuclear dyes after the transfection.
1. Cells are plated in two separate plates and prepared for transfection as described in Methods (Subheading 3.2). The cells
are then left untransfected (control), or transfected with 30 nM
of FITC-conjugated siRNA as described in Subheading 3.3.
2. After overnight incubation, the transfection medium is replaced
by fresh medium (see Subheading 3.3) in one of the two plates,
and the plate is incubated at 37C for a recovery period. For
the second plate, half of the transfection medium is carefully
removed and replaced by the same volume of culture medium
containing 10 g/mL of the nuclear dyes HO and PI (final
concentration of 5 g/mL cf. Subheading 2.3).
3. After 15 min incubation, half of the staining medium is carefully removed and replaced by the same volume of control
medium.
4. Cells are visualized and counted under a microscope with excitation wavelengths at:
(a) 358 nm (HO blue emission), for viable cells and early
apoptotic cells (25).
(b) 538 nm (PI red
(apoptotic + necrotic).
emission),
for
dead
cells
(c) 495 nm (FITC-conjugated siRNA green emission), for

transfected cells.
5. The efficiency of transfection is calculated as number of transfected cells/total number of cells (living + dead cells).
6. The toxicity of transfection is calculated as number of dead
cells/total number of cells (living + dead cells).
7. After 48 h of recovery, the second plate is removed from the
incubator and stained with HO/PI as described in above. The
viability of the cell is then evaluated under the microscope and
calculated as described in step 6.
187
The efficiency of transfection using fluorescent siRNAs should

be evaluated immediately after the overnight transfection period,
since there is a rapid decay in fluorescence. It is recommended, however, to evaluate cell viability in a second plate after a longer recovery
period (2448 h), in order to exclude later-stage toxicity of the
transfection. We have used this approach to define the best conditions to transfect INS-1E cells, primary rat beta cells and dispersed
human islets. Representative pictures of living (HO blue), dead
(PI red), and transfected (FITC green) INS-1E cells are shown
in Fig. 1a. The efficiency of transfection (calculated as described
above) was 95 3% for INS-1E cells and 81 3% for primary rat beta
cells (Fig. 1b, c). No significant cell toxicity was observed with this
transfection method (Fig. 1b, c). Of note, comparable results were
obtained in MIN6 mouse insulin-producing cells, in which the efficiency of transfection was 86 2%, with negligible toxicity (6 1% of
dead cells after transfection vs. 4 1% in the control).
For dispersed human islets, which contain a mixture of beta
cells (4060%) and other endocrine- and nonendocrine cells, we
performed immunostaining for insulin to evaluate the specific efficiency of transfection in the insulin-positive beta cell population.
The insulin staining for dispersed human islets is performed as
follows:
1. 2 104 cells/well are plated in sterile glass chamber slides and
transfected according to the procedure described above (see
Subheading 3.3).
2. After the transfection, cells are fixed with 100 L of 4% formaldehyde for 10 min at room temperature (RT).
3. Cells are washed 3 with PBS and permeabilized for 5 min
with 100 L/well of PBS/0.3% Triton-X 100 at RT.
4. Cells are washed 3 with PBS at RT and permeabilized for
5 min with 100 L/well of PBS/0.3% Triton-X 100.
5. After three washes with PBS at RT, cells are blocked with
100 L/well of blocking solution for 30 min at RT to decrease
nonspecific staining.
6. Diluted primary anti-insulin antibodies (100 L/well) are
applied for 1 h at RT.
7. After three washes with PBS, cells are incubated with 100 L/well
of the diluted secondary antibody for 60 min at room temperature and in the dark.
8. The cells are washed three times with PBS and nuclei are counterstained with 100 L/well of diluted HO (5 g/mL) for
15 min at room temperature and protected from light.
9. Cells are washed 3 with PBS, coverslips are mounted in
mounting medium, and immunofluorescence is visualized
under a microscope.
Fig. 1. Evaluation of siRNA transfection efficiency and toxicity in INS-1E cells, rat and human beta cells. (ac) INS-1E cells
or primary rat beta-cells were left untransfected or transfected with 30 nM of FITC-conjugated siRNA using the DharmaFECT
lipid reagent. Cells were labeled with the DNA-binding dyes Hoechst (5 g/mL) and PI (5 g/mL) immediately after transfection (plate 1 for evaluation of transfection efficiency) and after 48 h of recovery (plate 2 for evaluation of later stage
viability) and counted under a fluorescence microscope; (a) Representative pictures of the results observed in INS-1E cells
are shown; (b, c) The percentage of transfected fluorescence-positive cells (gray bars) or dead cells (black bars) for INS-1E
cells (b) or primary beta-cells (c) are shown. Results are mean SEM of five experiments. (de) Dispersed human islets
were left untransfected or transfected with 30 nM of FITC-conjugated siRNA using the DharmaFECT lipid reagent. Cells
were labeled with the DNA-binding dyes Hoechst (5 g/mL) and immunostained for insulin as described in Subheading 3.5.
Cells were then counted under a fluorescence microscope. The viability was evaluated in a separate plate using HO and PI
staining. (d) Representative pictures of the results are shown (the arrows indicate insulin- and FITC-positive cells). (e) The
percentage of transfected and insulin-positive cells (gray bars) or dead cells (black bars) are shown. Results are mean SEM
of three experiments.
189
Representative pictures of the insulin and FITC staining in

FITC-siRNA-transfected dispersed human islets are shown in
Fig. 1d. The efficiency of transfection for insulin-positive human
beta cells as assessed by this method was 87 2%, with no toxicity
(13 1% dead cells after transfection vs. 14.4 4% for untransfected
controls Fig. 1d, e). Of note, efficiency of siRNA transfection in
the human islet nonbeta cells was only 21 2%, indicating preferential transfection of beta cells under the present experimental
conditions.
3.6. Evaluation
of siRNA-Induced
Target Gene
Knockdown
Target gene knockdown mediated by transfection with siRNAs is

best evaluated by combining the determination of the target gene
mRNA and protein expression by respectively real-time quantitative RT-PCR and Western blot or other protein assays, such as
ELISA in the case of secreted proteins (e.g., chemokines). In the
case of transcription factors, inhibition of the target gene can be
additionally evaluated through the decrease of downstream genes.
In some cases, the absence of good antibodies for Western blot
poses a problem for validation; in these cases, we rely on mRNA
expression and determination of downstream genes/biological
effects (e.g., prevention of apoptosis in the case of a known proapoptotic gene).
An example of validation is shown in Fig. 2. INS-1E cells were
transfected with increasing amounts of either a siCtrl or a siRNA
targeting the iNOS as described above (Subheading 3.3). After
24 h of recovery, the cells were left untreated or challenged for
24 h with 100 U/mL interleukin-1, which induces iNOS expression and NO formation (26). SiRNA-mediated iNOS inhibition
was assessed at the mRNA and protein levels by respectively realtime RT-PCR and Western blot (Fig. 2a, b). Nitric oxide formation was assessed as nitrite accumulation in the cell culture
supernatant (Fig. 2c). The three methods confirmed a >75% inhibition of iNOS expression and nitrite in si iNOS-transfected
INS-1E cells, while cells transfected with the siCtrl showed similar
iNOS expression and nitrite formation as the nontransfected cells
exposed to IL-1.
3.7. Combining the Use

of siRNAs with
Microarray Analyses
and High-Throughput
Screening Methods
Due to its potency and ease of use, siRNA-mediated gene knockdown is a valuable tool to study in vitro different intracellular
mechanisms. We are presently combining the use of siRNA with
microarray analysis to investigate the gene networks responsible for
cytokine-induced beta cell apoptosis (10, 26). Thus, we have
recently performed microarray analyses of cells transfected with
control or siSTAT1 to identify the molecular pathways downstream
of the transcription factor STAT1 (27), a well-known regulator of
cytokine-induced beta cell apoptosis (28, 29). Since siRNA-mediated
gene knockdown induces a rapid and efficient inhibition of most
target genes in adult beta cells, it allows us to avoid putative
190
F. Moore et al.
Fig. 2. Evaluation of siRNA-induced target gene knockdown. INS-1E cells were left untransfected (NT) or transfected with
10, 30, or 50 nM of either control siRNA (siCtrl), or a siRNAs targeting iNOS (si iNOS). After 24 h of recovery, the cells were
challenged for 24 h with IL-1 (100 U/mL). (a) iNOS mRNA expression was assayed by RT-PCR and normalized for the
housekeeping gene GAPDH; (b) iNOS and -tubulin expression were evaluated by Western blot; (c) Nitrite concentrations
(nitrite is a stable product of NO oxidation) were evaluated in the in-culture supernatants using the Griess method. Results
are mean SEM of four experiments.
compensatory responses that are often observed with systemic

knockout models in which the target proteins are absent since fetal
life (these compensatory responses have been a particularly relevant
problem when dealing with KO mice targeting pro-inflammatory
cytokines). Another promising application of RNAi technology is
their use in high-throughput genome-scale loss-of-function screens
191
in cell culture systems. In this approach, hundreds to thousands of

siRNAs, constituting a siRNA library, are transfected individually
(one siRNA per well) into cells. The function of the target cells is
then analyzed in response to a given stimulus (e.g., interferons) by
the use of high-throughput evaluation methods. This approach has
been used for screens in both Drosophila melanogaster and mammalian cells (30), and genome-wide RNAi analysis have identified
novel components of the JAK/STAT (31) and p53 (32) pathways
in human cells. A present limitation of this approach is the elevated
cost of siRNA libraries, but it is expected that these costs will
decrease in the near future.
4. Notes
1. All siRNA should be reconstituted accordingly to the manufacturers instructions (usually in sterile ddH2O or siRNA
reconstitution buffer), aliquoted in sterile microtubes and snap
frozen at 80C. Once frozen, siRNA vials should not be
thawed/frozen more than two times to avoid degradation of
the duplexes and loss of activity.
2. Alternatively, OPTI-MEM may be replaced by any culture
medium in which the cells are usually cultured in, but the medium
should be free of any additives (serum, BSA, etc.).
3. In order to avoid toxicity, DharmaFECTs manufacturer
(Thermo Scientific) strongly recommends the use of antibioticfree medium before and during transfection. Respecting this
simple rule is crucial to obtain an efficient and nontoxic
transfection.
4. When setting the presently described experimental conditions,
we observed that BSA induces the aggregation of the lipid
reagent, nearly completely preventing transfection. Thus, the
medium used for transfection should be BSA-free.
5. We have tested up to now more than 40 siRNAs provided by
different manufacturers. Most of them were highly effective at
low concentrations (30 nM or less) and did not exhibit obvious off-target effects. A nonexhaustive list of web tools provided by suppliers for designing siRNA can be find in ref. 16.
6. When starting siRNA experiments, it is useful to test several
different sequence-independent control siRNAs to make sure
that it does not affect cellular function, viability, and/or metabolism. For example, we found that some commercially available control siRNAs alter insulin secretion in beta cells.
Moreover, significant toxicity may be observed with some
control siRNAs. We assume that this depends on the cell type.
192
F. Moore et al.
Of note, the control siRNA used in our experiments (see

Subheading 2.1, item 2) neither affects beta cell viability nor
modifies to major extent gene expression as evaluated by microarray analyses (27, and Fig. 2).
7. This step is very important and has a double goal: (1) To remove
the antibiotics from the medium to avoid toxicity associated
with the transfection; (2) To remove the BSA that is present in
the medium and which interferes with the transfection. Caution
should be taken when changing the medium, because the cells
have been attached for only 24 h and may easily detach. It is not
recommended to plate primary cells directly in antibiotic-free
medium to minimize the risk of infection.
8. When transfecting siRNAs, the concentration stated corresponds to the final siRNA concentration in the transfection
medium. According to the presently described protocol, this
means that the initial concentration of the siRNA will be
diluted 2 when adding the diluted DharmaFECT, and 5
more when medium is added to the siRNA/lipid complexes.
The initial dilution of the siRNA should be then 10 more
concentrated than the final desired siRNA concentration (e.g.,
transfection of 30 nM siRNA requires that the initial siRNA
dilution is 300 nM).
9. In our experimental settings, we observed that the ratio
between the concentration of DharmaFECT and the number
of cells to be transfected is not always linear. This may be also
affected by the cell type to be transfected. Thus, primary beta
cells require a higher concentration of DharmaFECT than
INS-1E cells for adequate transfection. When starting siRNA
experiments in a given cell type, it is recommended to establish
the best transfection conditions by testing different siRNA and
DharmaFECT concentrations.
10. We have tested transfection periods of 1418 h without noticing significant changes in the efficiency of transfection or
toxicity.
11. At this step, no toxicity was observed with medium containing
antibiotics; thus, antibiotic-free medium is not absolutely
required from this stage on.
12. When starting siRNA-based experiments to inhibit the expression of a given gene, we recommend to first establish a time
course of gene/protein expression 2496 h after siRNA transfection, in order to determine the required time to reach a
significant knockdown of the targeted protein. Of note, and
when dealing with fast proliferating cells, dilution of the siRNA
due to cell division may pose a problem after extended periods
of time.
11
Use of RNA Interference to Investigate Cytokine Signal Transduction...
193
Acknowledgments
This work has been supported by grants from the Fonds National
de la Recherche Scientifique (FNRS FRSM) Belgium, the
Communaut Franaise de Belgique Actions de Recherche
Concertes (ARC), the European Union (STREP Savebeta, contract no. 036903; in the Framework Programme 6 of the European
Community) and the Belgium Program on Interuniversity Poles of
Attraction initiated by the Belgium State (IUAP P6/40). F.M. is
the recipient of a Post-Doctoral Fellowship from FNRS, Belgium.
The authors have no duality of interest associated with this
manuscript. We thank M.A. Neef, G. Vandenbroeck, M. Urbain,
J. Schoonheydt, R. Leeman, A. M. Musuaya, and S. Mertens from
the Laboratory of Experimental Medicine, ULB, for excellent
technical support, Dr. Fernanda Ortis (Laboratory of Experimental
Medicine) for helpful comments and Dr. Piero Marchetti
(Department of Endocrinology and Metabolism, Metabolic Unit,
University of Pisa, Pisa, Italy) for providing the human islets used
for siRNA testing.
References
1. Fire, A., Xu, S., Montgomery, M.K., Kostas,
S.A., Driver, S.E., Mello, C.C. (1998) Potent
and specific genetic interference by doublestranded RNA in Caenorhabditis elegans.
Nature 391: 806811.
2. Ghildiyal, M., Zamore, P.D. (2009) Small
silencing RNAs: an expanding universe. Nat.
Rev. Genet. 10: 94108.
3. Naqvi, A.R., Islam, M.N., Choudhury, N.R.,
Haq, Q.M. (2009) The fascinating world of
RNA interference. Int. J. Biol. Sci. 5: 97117.
4. Elbashir, S.M., Harborth, J., Lendeckel, W.,
Duplexes of 21-nucleotide RNAs mediate
RNA interference in cultured mammalian cells.
Nature 411: 494498.
5. Wall, N.R., Shi, Y. (2003) Small RNA: can
RNA interference be exploited for therapy?
Lancet 362: 14011403.
6. Whitehead, K.A., Langer, R., Anderson, D.G.
(2009) Knocking down barriers: advances in
siRNA delivery. Nat. Rev. Drug Discov. 8:
129138.
7. Chen, Y., Stamatoyannopoulos, G., Song, C.Z.
(2003) Down-regulation of CXCR4 by inducible small interfering RNA inhibits breast cancer cell invasion in vitro. Cancer Res. 63:
48014804.
8. Cunha, D.A., Hekerman, P., Ladriere, L.,

Bazarra-Castro, A., Ortis, F., Wakeham, M.C.
Moore, F., Rasschaert, J., Cardozo, A.K.,
Bellomo, E., Overbergh, L., Mathieu, C., Lupi,
R., Hai, T., Herchuelz, A., Marchetti, P.,
Rutter, G.A., Eizirik, D.L., Cnop, M. (2008)
Initiation and execution of lipotoxic ER stress
in pancreatic beta-cells. J. Cell Sci. 121:
23082318.
9. Xia, H., Mao, Q., Paulson, H.L., Davidson,
B.L. (2002) siRNA-mediated gene silencing
in vitro and in vivo. Nat. Biotechnol. 20:
10061010.
10. Eizirik, D.L., Moore, F., Flamez, D., Ortis, F.
(2008) Use of a systems biology approach to
understand pancreatic beta-cell death in Type 1
diabetes. Biochem. Soc. Trans. 36: 321327.
11. Pipeleers, D.G., int Veld, P.A., Van de Winkel,
M., Maes, E., Schuit, F.C., Gepts, W. (1985)
A new in vitro model for the study of pancreatic
A and B cells. Endocrinology 117: 806816.
12. Rasschaert, J., Ladriere, L., Urbain, M.,
Dogusan, Z., Katabua, B., Sato, S., Akira, S,
Gysemans, C., Mathieu, C., Eizirik, D.L.
(2005) Toll-like receptor 3 and STAT-1 contribute to double-stranded RNA + interferongamma-induced apoptosis in primary pancreatic
beta-cells. J. Biol. Chem. 280: 3398433991.
194
F. Moore et al.
13. Ling, Z., Hannaert, J.C., Pipeleers, D. (1994)

Effect of nutrients, hormones and serum on
survival of rat islet beta cells in culture.
Diabetologia 37: 1521.
14. Lupi, R., Dotta, F., Marselli, L., Del Guerra, S.,
Masini, M., Santangelo, C. Patan, G.,
Boggi, U., Piro, S., Anello, M., Bergamini, E.,
Mosca, F., Di Mario, U., Del Prato, S.,
Marchetti, P. (2002) Prolonged exposure to
free fatty acids has cytostatic and pro-apoptotic
effects on human pancreatic islets: evidence
that beta-cell death is caspase mediated, partially dependent on ceramide pathway, and
Bcl-2 regulated. Diabetes 51: 14371442.
15. Delaney, C.A., Pavlovic, D., Hoorens, A.,
Pipeleers, D.G., Eizirik, D.L. (1997) Cytokines
induce deoxyribonucleic acid strand breaks and
apoptosis in human pancreatic islet cells.
Endocrinology 138: 26102614.
16. Pei, Y., Tuschl, T. (2006) On the art of identifying effective and specific siRNAs. Nat.
Methods 3: 670676.
17. Birmingham, A., Anderson, E., Sullivan, K.,
Reynolds, A., Boese, Q., Leake, D., Karpilow, J.,
Khvorova, A. (2007) A protocol for designing
siRNAs with high functionality and specificity.
Nat. Protoc. 2: 20682078.
18. Du, Q., Thonberg, H., Wang, J., Wahlestedt, C.,
Liang, Z. (2005) A systematic analysis of the
silencing effects of an active siRNA at all singlenucleotide mismatched target sites. Nucleic
Acids Res. 33: 16711677.
19. Reynolds, A., Anderson, E.M., Vermeulen, A.,
Fedorov, Y., Robinson, K., Leake, D., Karpilow, J.,
Marshall, W.S., Khvorova, A. (2006) Induction of
the interferon response by siRNA is cell type- and
duplex length-dependent. RNA 12: 988993.
20. Dande, P., Prakash, T.P., Sioufi, N., Gaus, H.,
Jarres, R., Berdeja, A., Swayze, E.E., Griffey,
R.H., Bhat, B. (2006) Improving RNA interference in mammalian cells by 4-thio-modified
small interfering RNA (siRNA): effect on
siRNA activity and nuclease stability when used
in combination with 2-O-alkyl modifications.
J. Med. Chem. 49: 16241634.
21. Hall, A.H., Wan, J., Shaughnessy, E.E., Ramsay,
S.B., Alexander, K.A. (2004) RNA interference
using boranophosphate siRNAs: structureactivity relationships. Nucleic Acids Res. 32:
59916000.
22. Zhang, N., Tan, C., Cai, P., Zhang, P., Zhao,
Y., Jiang, Y. (2009) RNA interference in mammalian cells by siRNAs modified with morpholino nucleoside analogues. Bioorg. Med.
Chem. 17: 24412446.
23. Editorial Comment (2003) Whither RNAi?

Nat. Cell Biol. 5: 489490.
24. Moore, F., Colli, M.L., Cnop, M., Esteve,
M.I., Cardozo, A.K., Cunha, D.A., Buglian,
M., Marchetti, P., Eizirik, D.L. (2009) PTPN2,
a candidate gene for type 1 diabetes, modulates
interferon-gamma-induced pancreatic beta-cell
apoptosis. Diabetes 58: 12831291.
25. Hoorens, A., Van de Casteele, M., Kloppel, G.,
Pipeleers, D. (1996) Glucose promotes survival
of rat pancreatic beta cells by activating synthesis of proteins which suppress a constitutive
apoptotic program. J. Clin. Invest. 98:
15681574.
26. Eizirik, D.L., Mandrup-Poulsen, T. (2001)
A choice of death the signal-transduction of
immune-mediated
beta-cell
apoptosis.
Diabetologia 44: 21152133.
27. Moore, F., Naamane, N., Colli, M.L.,
Bouckenooghe, T., Ortis, F., Gurzov, E.N.,
Igoillo-Esteve, M., Mathieu, C., Bontempi, G.,
Thykjaer, T., rntoft, T.F., Eizirik, D.L. (2011)
STAT1 is a master regulator of pancreatic beta
cell apoptosis and islet inflammation. J. Biol.
Chem. 286: 929941.
28. Callewaert, H.I., Gysemans, C.A., Ladriere, L.,
DHertog, W., Hagenbrock, J., Overbergh, L.
Eizirik, D.L., Mathieu C. (2007) Deletion of
STAT-1 pancreatic islets protects against streptozotocin-induced diabetes and early graft failure but not against late rejection. Diabetes 56:
21692173.
29. Gysemans, C.A., Ladriere, L., Callewaert, H.,
Rasschaert, J., Flamez, D., Levy, D.E., Matthys,
P., Eizirik, D.L., Mathieu, C. (2005) Disruption
of the gamma-interferon signaling pathway at
the level of signal transducer and activator of
transcription-1 prevents immune destruction
of beta-cells. Diabetes 54: 23962403.
30. Echeverri, C.J., Perrimon, N. (2006) Highthroughput RNAi screening in cultured cells: a
users guide. Nat. Rev. Genet. 7: 373384.
31. Muller, P., Kuttenkeuler, D., Gesellchen, V.,
Zeidler, M.P., Boutros, M. (2005) Identification
of JAK/STAT signalling components by
genome-wide RNA interference. Nature 436:
871875.
32. Berns, K., Hijmans, E.M., Mullenders, J.,
Brummelkamp, T.R., Velds, A., Heimerikx, M.
Kerkhoven, R.M., Madiredjo, M., Nijkamp,
W., Weigelt, B., Agami, R., Ge, W., Cavet, G.,
Linsley, P.S., Beijersbergen, R.L., Bernards, R.
(2004) A large-scale RNAi screen in human
cells identifies new components of the p53
pathway. Nature 428: 431437.
Chapter 12
Ligand Affinity Chromatography, an Indispensable Method
for the Purification of Soluble Cytokine Receptors
and Binding Proteins
Daniela Novick and Menachem Rubinstein
Abstract
Ligand affinity chromatography separation is based on unique interaction between the target analyte and
a ligand, which is coupled covalently to a resin. It is a simple, rapid, selective, and efficient purification
procedure of proteins providing tens of thousands fold purification in one step. The biological activity of
the isolated proteins is retained in most cases thus function is revealed concomitantly with the isolation.
Prior to the completion of the genome project this method facilitated rapid and reliable cloning of the
corresponding gene. Upon completion of this project, a partial protein sequence is enough for retrieving
its complete mRNA and hence its complete protein sequence. This method is indispensable for the isolation of both expected (e.g. receptors) but mainly unexpected, unpredicted and very much surprising binding proteins. No other approach would yield the latter. This chapter provides examples for both the
expected target proteins, isolated from rich sources of human proteins, as well as the unexpected binding
proteins, found by serendipity.
Key words: TBPII, IFNAR2, Soluble LDLR, IL-18BP, IL-32BP, PR3, Mass spectrometry,
Biomarkers, Interleukins, Interferons, Serendipity
1. Introduction
In 1968, Cuatracasas, Wilchek, and Anfinsen (1) coined the term
affinity chromatography in its form known today (2, 3). This
rapid and selective single-step purification procedure of proteins
exploits the immense power of bio-recognition between the covalently immobilized ligand to an insoluble matrix and the complementary target bio-molecule. Almost any given biomolecule has an
inherent recognition site through which it recognizes a partner
molecule. If one of these partners is immobilized on a polymeric
195
196
D. Novick and M. Rubinstein
carrier, it can be used to selectively capture the biomolecule of

interest. Isolation of a protein by affinity chromatography is a very
effective technique. It provides the protein in a fairly pure state and
enables its identification by partial sequencing, either by mass spectrometry or by N-terminal microsequencing. Upon completion of
the human genome project, a partial protein sequence is enough
for retrieving its complete mRNA and hence its complete protein
sequence.
Ligand affinity chromatography provides a comprehensive
solution for isolation and characterization of most receptors and
binding proteins. Any receptor or binding protein can be captured
on its immobilized ligand, provided that there is sufficient affinity
and a high specificity of interaction between the two partners. In
many cases, there is more than one type of receptor or binding
protein for a given ligand. Examples include the two membraneassociated TNF receptors (4) and the two membrane-associated
IL-1 receptors (5), each derived from a different gene. In other
cases, a single receptor gene may generate more than one mRNA
splice variant, yielding several receptor species (e.g., IFNAR2c, the
ligand-binding chain of Type I interferons and its short version
counterpart, the decoy receptor, IFNAR2b (57)).
In addition to membrane-associated receptors, circulating
ligand-binding proteins are quite common. One group of binding
proteins consists of soluble forms of cell surface receptors. Such
soluble receptors were identified for several hormones and growth
factors, including insulin, somatotropin (growth hormone), IGF,
and EGF. Among cytokine receptors, a soluble form of the IL-2
receptor (soluble Tac) was identified in cell culture supernatants
and then reported to be present in body fluids of normal individuals (8). These soluble receptors correspond structurally to the
extracellular ligand-binding domain of the cell surface receptor,
hence retaining their ligand-binding properties. Some of these
soluble receptors are produced by proteolytic cleavage of their cell
surface receptor counterparts, some from alternatively spliced
mRNA and some by both mechanisms (9). Studies performed in
our laboratory and in other laboratories revealed the presence of
many soluble cytokine receptors in body fluids, including the
receptors for IL-6, IFN-, TNF-, IL-1, IL-2, IL-4, IFN-/,
IL-13, IL-18, IL-22, IL-33, and others (1019). Based on these
studies, it became apparent that there are soluble receptors for
most if not all cytokines in the circulation, and that these receptors
play key roles in regulating cytokine-mediated biological activities
either in an antagonistic or an agonistic fashion.
Another group of circulating binding proteins is derived from
different genes and therefore is not homologous to the cell surface
receptors with which they share ligands. Examples of such proteins
include osteoprotegerin (20), Cytokine-like factor-1 (21), IL-18
binding protein isolated by us 10 years ago, (22) and the enzyme
12
197
proteinase 3 (PR3), recently described by us as an IL-32 binding

protein (23). The membrane bound receptor for IL-32 (24) is not
known till today. In an attempt to isolate this receptor via its soluble counterpart, we rather isolated what we named IL-32 binding
protein (IL-32BP). IL-32BP was isolated from normal human
urine and identified by a combination of ligand affinity chromatography and mass spectrometry. Eluted proteins from the affinity column were subjected to SDS-PAGE and the gel was stained with
mass spectrometry-compatible silver staining. Candidate bands
that were observed exclusively in the elution fractions and not
in the load or wash fractions were excised from the gel, the proteins were electro-eluted and digested with trypsin. The resulting
tryptic digest was subjected to liquid chromatography in tandem
with mass spectrometry (Smoler Protein Center, Technion, Haifa,
Israel). IL-32BP is not a soluble receptor but the well-characterized
proteolytic enzyme proteinase 3 (PR3) on the one hand and
an IL-32 binding protein on the other. PR3, also known as
Wegeners auto-antigen, is a neutrophil granule serine proteinase,
existing both in a soluble and in a membrane-bound form.
Autoantibodies to PR3, known as ANCA (anti neutrophil cytoplasmic autoantibodies) are the hallmark of Wegeners granulomatosis (25), and are present in systemic and small vessel autoimmune
vasculitic diseases. No other technique, but ligand affinity chromatography, would come across this protein and its functions, which
are deduced from the method used to purify it. First we demonstrated the enzymatic activity of the isolated PR3/IL-32BP. We
thus showed that limited proteolysis of IL-32 by this enzyme,
resulted in the formation of two peptides, a mixture of which
exhibited enhanced biological activity compared to the activity of
the intact cytokine. Employing Surface Plasmon Resonance
(Biacore, see Subheading 3.2.4) we demonstrated a high binding
affinity between PR3 and IL-32 (Kd in the nM range). In an
attempt to dissect the binding ability of PR3 from its proteolytic
activity, we treated PR3 with PMSF, a serine protease inhibitor,
prior to the binding studies. We showed that inactivation of PR3
with PMSF did not significantly reduce its binding affinity for
IL-32. This observation suggested that the binding of PR3 to the
immobilized IL-32 represents the non-enzymatic interaction of
enzyme to substrate and supports the concept that binding of IL-32
and the processing of IL-32 are two separate properties of PR3.
Ligand affinity chromatography has been successfully employed
for isolating and identifying the above mentioned soluble receptors and binding proteins. In some cases, isolation of these
binding proteins and their N-terminal amino acid sequencing
enabled cloning or identification of their unknown genes at that
time (6, 11, 22).
This chapter describes the methods used in our laboratory for
the purification, characterization, and molecular cloning of soluble
198
cytokine receptors, binding proteins, and cell surface receptors.

The immense power of ligand affinity chromatography in protein
purification is demonstrated when 1,000-fold purification is
achieved in one step. For example, for the isolation of cell surface
receptors, we used 1011 cells per batch, or two term placentas per
batch as rich sources of proteins. For the isolation of soluble
receptors and binding proteins we used highly concentrated
(5001,000-fold) urinary proteins (500 mL/batch equivalent to up
to 500 L of human urine; ca. 100 mg protein/L). In the case of
the latter, the affinity column load fraction contained 2550 g of
protein mixture while the elution fraction contained ca. 0.25 mg
of proteins, including the target soluble receptor or binding protein. Thus ca. 100,000-fold purification of the target protein was
achieved in one step. Specific ELISAs confirmed no big loss of
specific protein in this purification procedure. Our approach of isolating proteins by ligand-affinity-chromatography enabled rapid
and efficient isolation of seven soluble receptors corresponding to
cell-associated receptors (examples in Fig. 1ad). Moreover, this
method also enabled the isolation of non-receptor binding proteins and associated enzymes (Fig. 1e, f). No other approach
would yield the latter. Thus with the availability of a pure and active
ligand on the one hand, and a rich source of proteins on the other,
ligand affinity chromatography virtually guarantees a successful
target-protein isolation (42).
2. Materials
2.1. Receptor Isolation
1. Affi-Gel 10, or any other suitable pre-activated carrier for

ligand immobilization (Bio-Rad, Hercules, CA).
2. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM
KCl, 1.5 mM KH2PO4, 6.5 mM Na2HPO4, pH 7.4, if necessary, 0.02% (w/v) NaN3.
3. Benzamidine.
4. 0.5 m membrane (e.g., Pellicon Cassette, Millipore, Bedford,
MA).
5. 10 kDa cutoff membrane (PTGC 10, Millipore).
6. Hypotonic buffer: 10 mM TrisHCl, pH 7.5, 1 mM MgCl2,
1 mM CaCl2, 1 mM phenylmethylsulfonyl fluoride (PMSF),
22 TIU/mL aprotinin. Additional protease inhibitors such as
pepstatin and leupeptin may be included.
7. Vortex shaker.
8. Tris-sucrose buffer: 20 mM TrisHCl, pH 7.4, 0.25 M sucrose,
1 mM PMSF.
9. Ultratorax homogenizer.
12
199
Fig. 1. Various soluble receptors purified from concentrated human urinary proteins by ligand affinity chromatography. (a).
Lane 1: Molecular mass (kDa) markers. Lane 2: Urinary proteins purified by ligand (IL-6) affinity chromatography followed
by HPLC and analyzed by SDS-PAGE and silver staining (10). (b) Lane 1: Molecular mass (kDa) markers. Lane 2: Ligand
(IFN-) affinity purified urinary proteins (300 ng) analyzed by SDS-PAGE and silver staining (10). (c) Lane 1: Urinary proteins
and Lane 2: ligand (IFN-2) affinity purified urinary proteins analyzed by SDS-PAGE and silver staining (6). Molecular mass
markers (kDa) are indicated on the left side. (d) Immunoblot analysis of cellular IFNAR2 with anti IFNAR2 antibodies. Lane
1: A detergent-extract of Daudi cells (108), purified on monoclonal antibodies to soluble IFN/ receptor. Lane 2: Crude
detergent extract of Daudi cells (5 106). Lane 3: Crude detergent cell extract of mouse NIH 3T3 cells (5 106) (6). (e, f)
Two binding proteins purified from concentrated human urinary proteins by ligand affinity chromatography and analyzed
by SDS-PAGE and silver staining. (e) Lane 1: Crude urinary proteins. Lane 29: Elution fractions (18) from the IL-18
column. Lane m: Molecular mass markers (kDa). The arrow indicates the IL-18 binding protein (22). (f) Lane 1: Wash fraction
representing crude urinary proteins. Lanes 26: Elution fractions (15 respectively) from the IL-32 column. Lane 7:
Molecular mass (kDa) markers indicated on the right. The arrow indicates the IL-32-binding protein (23).
10. Gauze.
11. HEPES buffer: 50 mM HEPES, pH 7.5, 1 mM PMSF,
20 TIU/mL aprotinin and other anti-proteases, and 0.02%
(w/v) NaN3.
12. Liquid nitrogen.
13. Solubilization buffer: 1% Triton X-100, 10 mM HEPES, pH
7.5, 150 mM NaCl, 1 mM PMSF, 20 TIU/mL aprotinin, and
other anti-proteases.
14. Low pH buffer: 25 mM citric acid, pH approximately 2.5,
1 mM benzamidine, 0.01% (w/v) NaN3.
200
15. High pH buffer: 25 mM Na2CO3, pH approximately 11.0,

0.53 M NaCl, 1 mM benzamidine, 0.01% (w/v) NaN3.
16. 1 M Na2CO3.
17. 1 M TrisHCl, pH 9.5.
18. Acetic acid (3 M and 5%, v/v).
2.2. Receptor
Characterization
1. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis

(SDS-PAGE) equipment.
2. Fixing solution: 50% (v/v) methanol, 10% (v/v) acetic acid.
3. 10% (v/v) glutaraldehyde.
4. Ethanol.
5. 30% (w/v) NaOH.
6. Ammonia.
7. Silver nitrate.
8. 1% (w/v) citric acid.
9. 37% (v/v) formaldehyde.
10. Acetone.
11. PD-10 columns (Amersham-Pharmacia, Piscataway, NJ).
12. Gelatin.
13. 0.5 M phosphate buffer, pH 7.5.
14. Chloramine-T.
15. Na[125I].
16. NaHSO3.
17. KI.
18. Bovine serum albumin (BSA), 1% (w/v) in water.
19. 20% (w/v) trichloroacetic acid (TCA).
20. Disuccinimidyl suberate (DSS).
21. Dimethyl sulfoxide (DMSO).
22. Stop solution (cross-linking): 1 M TrisHCl, pH 7.5, 1 M
NaCl.
3. Methods
3.1. Receptor Isolation
Procedures
3.1.1. Preparation
of a Ligand Affinity
Column
Successful receptor isolation depends on availability of a pure ligand

(see Note 1), which retains its binding capability upon immobilization to an affinity resin. Affinity resins suitable for chromatography
of proteins usually consist of derivatized agarose. There are several
types of chemistries and spacers of various lengths that may be useful to avoid steric hindrance. Pre-activated carriers are more convenient
to work with. Several suppliers provide such carriers and also
12
201
include detailed procedures for ligand immobilization and elaborate

on performance of the chromatographic procedure (AmershamPharmacia, Bio-Rad, Sigma, Pierce and others).
The following procedure uses Affi-Gel 10 (Bio-Rad, Hercules,
CA, USA). This carrier is based on the active N-hydroxysuccinimide
ester, which is relatively stable and reacts rapidly with primary (and
secondary) amines of proteins, such as N-terminal sites and -amino
groups of Lysine. Affi-Gel 10 is suitable for coupling proteins having isoelectric points from 6.5 to 11. Proteins with isoelectric
points below 6.5 are better coupled to Affi-Gel 15.
1. Couple 25 mg (at a concentration of at least 1 mg/mL) of
highly purified (preferably homogenous) ligand (see Note 2)
to Affi-Gel 10 (0.51 mL wet beads) according to manufacturers instructions. Make sure that the protein solution does
not contain primary amines such as TRIS, glycine, cell culture
media, or ammonium salts. Usually, about 8090% of the
ligand is immobilized.
2. Store the resulting gel at 4C in phosphate-buffered saline
(PBS) containing 0.02% NaN3. Stability of the immobilized
ligand upon repeated use of the column is not predictable. In
most cases it may be used for chromatography numerous
times.
3. Test the stability of the ligand by a bioassay following treatment with high and low pH. This will enable choice of the
most suitable elution conditions. Elution at a pH value that
inactivates the ligand may still be employed for small-scale procedures, provided that the gel is rapidly neutralized following
elution.
3.1.2. Preparation of Crude
Receptor Extracts
Successful receptor isolation depends not only on the availability of

a pure ligand but also on a suitable and sufficient protein source
(see Note 3) of the receptor to be isolated. Sufficient amounts will
ensure a final yield of the isolated receptor, enabling its characterization (by e.g. N-terminal amino acid sequencing, mass spectrometry, SDS-PAGE, evaluation of biological activity, etc.) and antibody
generation (see Note 4). Isolation of membrane-bound receptors requires a preliminary step of membrane isolation, followed
by solubilization with a detergent. A number of mild detergents
may be employed in the solubilization of cytokine receptors.
Triton-X-100 is an example for a non-ionic detergent; CHAPS,
which is more effective for membrane solubilization, is an example
of a zwitterionic detergent.
Isolation of soluble receptors does not require the use of detergents and hence is simpler to perform. Soluble receptors may be
isolated from cell culture supernatants, from plasma, serum, or
urine. Urinary proteins are derived from those plasma proteins that
pass through the kidney. Because the kidney retains high molecular
202
weight proteins more effectively, the urine of healthy individuals is

enriched with proteins <65,000 Da, including most soluble receptors and binding proteins. The low proportion of very high molecular weight proteins in the urine facilitates its concentration and
makes it a convenient source for isolation of serum proteins with a
molecular weight below 65,000 Da.
3.1.2.1. Preparation
of Crude Urinary Proteins
1. Collect urine from healthy donors in the presence of a protease

inhibitor (e.g. 1 mM benzamidine) and 0.02% NaN3. Store at
4C (all procedures should be done at 4C to minimize
degradation).
2. Filter pools of urine (at least 100 L per batch) through a 0.5 m
membrane (e.g. Pellicon Cassette, Millipore, Bedford, MA,
USA).
3. Concentrate the clear filtrate 5001,000-fold by tangential
ultrafiltration, using a membrane of 10,000 Da cutoff (e.g.,
PTGC 10 membrane, Millipore). Remove the remaining small
molecules by washing with PBS containing 0.02% NaN3 and
1 mM benzamidine.
4. Keep the concentrated preparation of urinary proteins frozen
at 20C.
3.1.2.2. Cell Culture

Supernatants
Cell culture supernatants may be concentrated using the same

procedure as used for the concentration of urinary proteins (26, 27).
However, cells must be quickly removed, e.g., by centrifugation, as
they tend to release proteases. Also, the presence of serum (e.g.,
fetal bovine serum) limits the degree of concentration, as it is not
practical to work with serum concentrations exceeding 100%.
Therefore, if possible, it is recommended to use a minimal amount
of serum as a cell culture additive or employ serum substitutes.
3.1.2.3. MembraneAssociated Receptors

from Cultured Cells (11)
Perform all procedures at 4C.

1. Wash cells (at least 3 1010 per batch) with PBS. Collect the
cells by centrifugation (1,500 g, 7 min).
2. Disperse the cell pellet and suspend the cells in hypotonic
buffer.
3. Mix vigorously on a Vortex shaker and allow standing for
5 min.
4. In order to clarify the lysate and remove debris, spin at three
different stages: 700 g for 5 min (discard pellet), 3,500 g for
10 min (discard pellet), and 40,000 g for 1 h. Collect the
pellet from the last spin.
5. Solubilize membranes as described
Solubilization of membranes.
in
Subheading
12
3.1.2.4. MembraneAssociated Receptors
from Tissues
203
This procedure is a modification of the procedure described by

Hock and Hollenberg (28). We successfully used it with human
term placenta.
1. Separate a fresh placenta (approximately 350 g) from its amniotic
sac and cut into pieces of about 20 g.
2. Mince the pieces by a meat grinder into saline and wash thoroughly with saline.
3. Add two volumes of TrisHCl-sucrose buffer pH 7.5 and homogenize by e.g., Ultratorax homogeniser (5 30 s, setting 5).
4. Filter the homogenate through two layers of gauze and spin
the filtrate (600 g, 10 min) to remove debris.
5. Clarify the supernatant by centrifugation at 10,000 g for
30 min.
6. Adjust the supernatant to a final concentration of 0.1 M NaCl
and 0.2 mM MgSO4 and spin at 48,000 g for 40 min. Collect
the pellet.
7. Suspend the pellet in an equal volume of HEPES buffer.
8. Freeze quickly in liquid nitrogen.
9. Solubilize membranes as described under Subheading
Solubilization of membranes.
One average-size placenta yields ca. 450 mg of membranes
(30 mg/mL).
3.1.2.5. Solubilization
of Membranes
1. Suspend either intact, PBS-washed cells (30 mL of cell pellet)

or cell membranes (30 mg/mL, 15 mL) in an equal volume of
a solubilization buffer.
2. Incubate for 1 h with occasional shaking.
3. Centrifuge (10,000 g, 15 min).
4. Collect the supernatant and clarify by ultracentrifugation
(100,000 g, 4C, 60 min).
5. The resulting supernatant is ready for ligand affinity
chromatography.
3.1.3. Ligand Affinity

Chromatography
Perform all steps at 4C. The procedure described is optimized for

soluble receptors. For solubilized membrane-bound receptors add
0.1% (v/v) Triton-X-100 or another detergent of choice to all the
buffers.
1. Wash the affinity column with PBS.
2. Spin (10,000 g, 15 min) the crude preparation containing
soluble or solubilized receptor (the Load fraction) prior to
loading on the column. Set aside an aliquot for assays.
3. Apply the Load fraction (see Note 5) to the column at a flow
rate of 1520 mL/h.
204
4. Collect the effluent fraction and keep for assays.

5. Wash the column (at a gravitation rate) with 250 column
volumes of PBS containing 0.51 M NaCl. Collect the first
and the last 10 column volumes separately for assays.
6. Elute the receptor with either low or high pH buffer. Low pH
buffer was found in most cases to be more effective than high
pH buffer. If the ligand is sensitive to low pH, elution at a high
pH should be tried. In such cases, elution may be improved by
the addition of high salt (13 M).
Collect ten fractions of one column volume each.
7. Immediately neutralize the elution fractions (e.g. with 1 M
Na2CO3 or with 1 M TrisHCl pH 9.5, in the case of the low
pH elution buffer and with 3 M acetic acid in the case of the
high pH elution buffer) (see Note 6). Calibrate the neutralization volumes needed to neutralize the elution buffer of choice:
check the pH of the first three elution fractions (use 1 L on
pH indicator paper) and note that most probably the first elution fraction will be at neutral pH, the second will need half
the neutralization buffer volume and only the third will need
the entire volume as determined by the calibration.
8. Wash the column immediately with PBS and store in PBS containing 0.02% (w/v) NaN3.
9. Determine the protein content in all the samples from the various purification steps. Alternatively, when the amount of pure
protein is limited, protein content can be estimated following
SDS-PAGE and silver staining of the gel.
10. Keep eluted fractions at 4C. Repeated freezing and thawing is
not recommended. Stability studies should be performed for
the determination of the best storage conditions.
3.1.4. Additional
Purification Steps
In most cases an additional step of purification is required for

obtaining the receptor in sufficient purity for further characterization. If the contaminating proteins differ very much in size from
the protein of interest, they may be resolved by size exclusion chromatography (using e.g. Superose 12, or Superdex 75, AmershamPharmacia) (5). As the volume loaded on size exclusion columns is
limited, eluted fractions from the ligand affinity chromatography
column should first be concentrated. Ultrafiltration in small devices
(e.g., Ultrafree MC cutoff 10 kDa, Millipore) is recommended to
minimize losses by adsorption to walls.
Alternatively, proteins may be further resolved following ligand
affinity chromatography by reversed-phase (RP) HPLC (29). Not
all proteins will resist the rather harsh conditions of RP-HPLC and
some of them may not elute well from the column. However, the
very high resolution of RP-HPLC makes this procedure very
attractive. Use columns that have the largest pore size possible
(100300 ). Octyl (C8) chains offer a good compromise between
columns with excessive hydrophobicity (C18) and those exhibiting
12
205
poor coverage of the silica surface (C4). Use gradients of either

acetonitrile or 1-propanol as the organic modifier and 0.10.3%
trifluoroacetic acid (TFA). The organic solvent should be removed
immediately after completion of the fractionation by evaporation
(e.g., vacuum centrifugation) and residual TFA neutralized with a
mild base, e.g., triethanolamine. We used Aquapore RP-300,
4.6 30 mm cartridge columns (Brownlee Labs, Perkin Elmer,
Norwalk, CT) to further resolve the soluble IL-6 receptor (10) as
well as many other proteins.
3.2. Receptor
Characterization
Procedures
3.2.1. SDS-PAGE
3.2.2. Silver Staining

of Polyacrylamide Gels
As a routine, run SDS-PAGE with the tested samples both under

reducing and non-reducing conditions (e.g. DTT to a final concentration of 25 mM). Proteins isolated from urine are expected
to be smaller than 65 kDa, thus 10% or 12% acrylamide gels are
recommended. Membrane-associated receptors are usually larger
than 65 kDa, requiring 7.510% acrylamide gels to achieve a satisfactory separation. Include in your gel samples from each purification step diluted according to the method of staining to be used.
Optimal silver staining of the gel is achieved if the protein content
does not exceed 100 ng or 200 ng per band. If the protein content of the elution fractions was not determined, load aliquots of
up to 60 L/lane. Always include in your gel a control lane containing your sample buffer, since the reducing agent present in the
sample buffer tends to give artifacts at around 67 kDa upon silver
staining (30).
Though many commercial kits and silver staining protocols exist in
the market, the following protocol (31) had proven to be the best
in our hands for a sensitive detection of proteins:
Note: Do not use autoclaved water, since the AgNO3 might
precipitate. Use highly purified double distilled water (e.g., from a
Milli-Q water purification system; Millipore, Bedford, MA, USA)
to prepare all your solutions and for all the washings of your gel.
1. Fix the gel overnight in methanol (50%, v/v), acetic acid (10%,
v/v) solution (200 mL).
2. Wash the gel (four times, 15 min each) with water.
3. Fix the gel with glutaraldehyde (10%, v/v, 125 mL) for 30 min.
Note that glutaraldehyde commercial stock is 25%.
4. Wash the gel (six times, 10 min each) with water.
5. Freshly prepare a silver stain solution for 12 mini gels as
follows:
Water
(105 mL)
Ethanol
(14.7 mL)
NaOH (30%)
(0.25 mL)
NH4OH
(1.5 mL)
206
Weigh 0.8 g AgNO3 (Analytical grade) and dissolve in

4 mL water. Add the silver solution dropwise with stirring into
the above mixture. Keep in the dark until used.
6. Soak the gel in the silver stain solution (10 min).
7. Wash the gel with water four times, 15 min each wash.
8. Develop the gel (530 min) in the following freshly prepared
developer solution:
Water
(179 mL)
Ethanol
(20 mL)
Citric acid (1% stock)
(1 mL)
Formaldehyde (37%)
(0.1 mL)
9. Stop the developer by brief washing with water (twice). Some

protocols suggest a final rinse in water with 5% (v/v) acetic
acid.
Use a glass container rinsed well in highly purified water and
then in acetone. Use gloves without powder and do not touch the
gel with your fingers.
All steps should be performed on a shaker to prevent the gel
from sticking to the glass container.
3.2.3. Cross-Linking
of the Purified Protein
to Its Ligand
3.2.3.1. Labeling
of Ligands with (125I)
The simplest and most effective way to label proteins is by the

Chloramine-T technique (32), which labels proteins at their
tyrosine residues. If the protein is sensitive to oxidation, the Bolton
and Hunter technique (33), based on acylation of amine residues,
provides gentler labeling conditions.
In order to minimize any oxidation-induced damage to ligandbinding capability, the following modification of the Chloramine-T
method should be used. All solutions are freshly prepared at room
temperature and the labeling is performed on ice. Labeling is more
efficient if the protein concentration is higher than 100 g/mL
(preferably 1 mg/mL).
1. Equilibrate a disposable size exclusion chromatography column (510 mL bed volume, e.g., pre-packed Sephadex G-25,
PD-10 column, Amersham-Pharmacia) with 25 mL PBS containing 0.25% (w/v) gelatine, 0.02% (w/v) NaN3 (equilibration buffer).
2. Mix 125 g of protein in a volume of 25100 L PBS (or
another buffer compatible with the Chloramine-T technique)
with an equal volume of 0.5 M phosphate buffer pH 7.5 [A].
3. Mix 25 L Chloramine-T solution (1 mg/mL in H2O) with
1 mCi of Na[125I] in 10 L for 20 s [B].
4. Add [A] to [B] and incubate for 20 s.
12
207
5. Stop the reaction by the addition of 50 L of NaHSO3 (5 mg/mL,

50 mL) and KI (5 mg/mL, 50 L) for 2 min.
6. Separate free from bound iodine using the size exclusion column. For optimal separation, let the column run until the bed
is drained, apply the iodination mixture, allow the column to
drain, add 1 mL equilibration buffer, let drain, and repeat five
more times. Collect six fractions of 1 mL each into polypropylene tubes. The iodinated protein elutes in fractions 3 and 4
(monitor with Geiger counter).
7. Count 2 L from each fraction in a gamma counter. Typically
almost no counts are observed in the first two elution fractions.
Up to 28 105 cpm/L are obtained in fractions 3 and 4.
Fractions 5 and 6 contain no more than 5% of the labeled
protein.
8. Store the radiolabeled ligand at 4C.
Check the extent of iodinated protein by TCA precipitation:
1. Mix 2 L of each of the eluted fraction No. 3 and 4 with
100 L of 1% (w/v) BSA, add an equal volume of 20% (w/v)
TCA and incubate for 20 min on ice.
2. Spin at 13,000 g for 2 min.
3. Separate the supernatant from the precipitate and read both in
a gamma counter.
Determine the counts in the precipitate as a percentage of the
total counts in the 2 L sample tested. Usually over 90% are TCA
precipitable, providing a labeling level of ca. 108 cpm/g.
3.2.3.2. Cross-Linking
of a Labeled Ligand
to the Purified Soluble
Receptor
This procedure is performed on ice. Avoid buffers containing free

amines (e.g. culture media, Tris, glycine).
1. Mix an aliquot (40 L) of the affinity purified receptor peak
fraction (chosen according to the silver staining of the gel)
with the iodinated ligand (0.55 106 cpm) for 1 h.
2. Add freshly prepared disuccinimidyl suberate (DSS, Pierce,
20 mM, dissolved in DMSO and kept at room temperature) to
a final concentration of 12 mM. Incubate the mixture for
20 min on ice (see below for considerations for choice of crosslinkers).
3. Stop the reaction with a solution of 1 M TrisHCl, pH 7.5,
1 M NaCl to a final concentration of 100 mM.
4. Add sample buffer (containing a reducing agent) and load on
a 7.5% SDS-PAGE.
Alternatively, the cross-linked product can be immunoprecipitated or immunoaffinity purified (34) using specific antibodies to
the ligand and then loaded on the gel.
208
There are many types of cross-linkers, which are comparable in

their efficiency. Cross-linkers may be either water-soluble or insoluble. In terms of chemistry, two groups were used successfully, the
activated N-hydroxysuccinimide esters and the imidoesters.
Imidoesters (e.g., dimethyl suberimidate.HCl) are water-soluble
and relatively inexpensive. The simple N-hydroxysuccinimide esters
such as DSS need to be dissolved in a water-miscible solvent
(DMSO, DMF, dioxan, etc.). Some of the N-hydroxysuccinimid
esters are available as water-soluble sulfonated molecules. However,
these are much more costly. All cross-linkers are sensitive to humidity, therefore should be stored under desiccation and also need to
be dissolved immediately before use.
3.2.4. Determining
the Affinity of a Soluble
Receptor to Its Ligand
To date, most of the soluble receptors identified were found to

function as inhibitors. At lower concentrations these same soluble
receptors may function as stabilizers of the ligand as was reported
for IL-4 and TNF- (35, 36). An example for an exception is the
soluble IL-6 receptor, which acts as an agonist (34). IL-6 binds to
the circulating soluble IL-6 receptor rather than to its membranebound form receptor, and the complex binds to the gp130, the
signal transducing chain of the IL-6 receptor complex (11, 37).
This agonistic feature of the soluble IL-6 receptor enables IL-6
signaling in cells lacking the ligand-binding chain of the IL-6 receptor complex. High affinity binding of the ligand with its receptor
and slow off rate of this binding strongly indicates that the
receptor acts as an inhibitor rather than a carrier protein. The
IL-18 Binding Protein (22) is an example of such a unique high
affinity (Kd = 0.4 nM) soluble receptor with an association rate constant (1.38 106/M/s) and a markedly reduced dissociation rate
constant (6.43 104/s), thus inhibiting the biological activity of
its ligand at equimolar ratios (38).
Surface plasmon resonance (BIAcore, Uppsala, Sweden) measures efficiently the association and dissociation rate constants and
hence the affinity of the soluble receptor to its ligand. This method
is based on the use of disposable chips containing a gold leaf layer,
to which activated dextran is attached on one side. The chip is
placed in the BIAcore system, which couples a microfluidic system
to the dextran side and applies a polarized light beam to the gold
side. Changes in the mass of proteins attached to the dextran side
are accurately measured as changes in the angle of surface plasmon
resonance (angle of no reflection).
To perform the analysis, one protein is immobilized onto the
BIAcore chip while the counter protein is analyzed at several
concentrations. In our hands, immobilizing soluble receptors
rendered them inactive, whereas immobilized cytokines retained
their binding properties. The BIAcore system is very sensitive,
enabling analyses to be performed at protein concentrations
around 1 g/mL.
12
209
1. Immobilize the ligand to an individual channel in a BIAcore

sensor chip as recommended by the manufacturer. Usually low
levels of binding (5001,000 resonance units) are suggested to
minimize errors due to re-binding at the dissociation phase.
2. Perform binding-dissociation experiments at several concentrations of the soluble receptor as recommended by the manufacturer. Analyze the binding constants and kinetics of binding
of the soluble receptor preparation to its ligand accordingly.
3.2.5. Bioassay
of the Soluble
or Solubilized Receptor
The goal of protein purification is to identify its function. Therefore,

demonstrating the biological activity of an isolated protein is an
essential step in this process. Because biological activity is a very
broad term, a variety of procedures are used for its measurement.
Generally speaking, bioassays of soluble receptors are based on
their ability to modulate the biological activity of their respective
ligands. In vitro bioassays using cell cultures are essential due to
the limited yield of the isolated receptor. Analysis with the BIAcore
system may also be used for determining whether the soluble
receptor retains its ligand-binding activity, but only a bioassay will
show if the soluble receptor acts as an antagonist, competing with
the cell-surface receptor, or as an agonist, e.g., in the case of the
soluble IL-6 receptor.
A simple bioassay with a reliable read-out should be chosen.
Examples include the inhibition of the antiviral activity of IFNs
in vitro (6), inhibition of the induction of IFN- by IL-18 for measuring IL-18BP (22), or inhibition of cytotoxicity to cells by soluble receptors for TNF (TBPI and TBPII) (11, 27). See Note 6 for
handling of samples to be tested in a bioassay.
3.2.6. Protein Sequence

Analysis (by Edman
Degradation and/or Mass
Spectrometry)
In most of our studies, sequence analysis of isolated proteins was

performed by N-terminal sequencing using the Edman degradation procedure. This procedure provides the sequence of amino
acids comprising the N-terminus of the isolated protein or its fragments. About 110 pmol of protein or peptide may be analyzed by
modern protein microsequencers. For analysis, the protein should
be either purified to homogeneity or resolved on SDS-PAGE followed by electroblotting onto a PVDF membrane and staining. As
a rule, sequencing proteins from solution is more efficient. Protein
fragmentation (CNBr, trypsin, etc.) may be used if the N-terminus
is blocked or additional sequence data is required. More recently,
with the completion of the genome project mass spectrometry
became the method of choice for sequencing the isolated proteins,
since all potential human protein sequences are contained in the
database. Mass spectrometry is more sensitive than N-terminal
sequencing enabling analysis at sub-picomole levels. The disadvantage of this method is that is not quantitative and yields the
sequence of all peptides present in the tested sample.
210
3.3. Further Structural

Characterization
of the Isolated
Receptor
A partial protein sequence of an isolated receptor is sufficient for

identification of the complete sequence of its various variants using
the BLAST tool of NCBI (http://www.ncbi.nlm.nih.gov). BLAST
also identifies related proteins and homologous proteins in other
species. For example following isolation of human IL-18BP and its
partial N-terminal sequence, extensive use of such databases and
tools facilitated its cloning, led to identification of the murine
IL-18BP as well as a family of pox-virus encoded proteins that bind
IL-18 and block its biological activity (22, 39). Such virus-encoded
soluble decoy proteins, that enable the virus to evade the immune
system of the host, were described for Type I and Type II interferons, for TNF, and for other cytokines (40).
Searching genomic databases is just the first step. In many
cases, a single gene generates numerous mRNA splice variants,
whose relevance to biologically active proteins needs to be determined empirically. This requires complete sequencing of cDNA
clones, either isolated from cDNA libraries or obtained as individual clones from public collections (e.g., http://www.ATCC.org).
RT-PCR and RNA blot hybridization as well as Western blot analysis must be used to indicate which of the mRNA molecules predicted to exist by the isolated cDNA clones is actually expressed.
The presence of a particular mRNA species does not serve as a
proof that its corresponding protein is actually translated in vivo
and is biologically active. Therefore, this protein must be expressed
in order to compare its biological activity to that of the originally
isolated protein. A good case in point is that of IFNAR2, the
ligand-binding subunit of the Type I IFN receptor. Three mRNA
splice variants were identified: IFNAR2a, coding for a soluble
IFNAR2, IFNAR2b, coding for a transmembrane receptor that
has a short cytoplasmic domain (6) and IFNAR2c, coding for a
transmembrane receptor that has a longer cytoplasmic domain (7).
RNA blot analysis revealed that IFNAR2b was the major species at
the mRNA level. However, the corresponding protein was barely
detectable by immunoblotting and the recombinant protein bound
IFNs but was unable to transduce the signal and thus failed to
induce an antiviral response (41). Therefore, IFNAR2b can be
defined as a decoy receptor. As it turned out, the minor mRNA
species IFNAR2c coded for the biologically fully active receptor
subunit, whereas very little IFNAR2b protein is found in cells
despite the high abundance of its mRNA (7).
In addition to the initial identification of phenotypic expression and protein characterization, a key parameter in proteomic
analysis is the ability to quantify proteins of interest. Quantitation
remains a vital analytical component of target validation for proteomic analysis, and of the determination of translational effects
that affect protein production and function. Unfortunately, mass
spectrometric signals alone may not always correlate precisely with
the amount of the analyte present in a given sample.
12
3.4. Preparation
of Antibodies
211
It is highly recommended to generate polyclonal and monoclonal

antibodies to the purified proteins as soon as sufficient amount of
homogenous protein is available. Note that 510 g of pure protein
is enough for one injection of a rabbit and 15 g is sufficient for one
injection of a mouse. A good titer is obtained after four such injections. Such antibodies facilitate further characterization of the purified
protein employing immunostaining, immunoblotting, and ELISA.
4. Concluding
Remark
In summary, our approach of combining an enriched source of
proteins together with highly specific isolation method, the ligand
affinity chromatography, enabled us rapid and efficient isolation of
not only soluble receptors corresponding to cell-associated receptors but also independent binding proteins and associated enzymes.
No other approach would yield the latter. During the past 20 years,
we isolated seven such soluble receptors (IL-6R, IFN-R, TBPI,
TBPII, LDLR, IFN/R, IL-1R Type II) and three unexpected
binding proteins, the IL-18BP, the IL-32BP/PR3 (42) and a heparanase binding protein.
We showed that ligand affinity chromatography is an indispensable method for the isolation of expected but mainly unexpected, unpredicted, and very much surprising interacting proteins.
The first surprise we encountered in 1989 when it was this method
that yielded not one but two soluble TNF receptors (soluble TBPI
and TBPII) (11). The TBPII would have never been discovered by
the multi-step chromatographic procedure used to purify the TBPI
(27). TBPII derivative had been developed to a drug known as
Enbrel and is used by over 500,000 rheumatoid arthritis and
rheumatoid psoriasis patients. In 1994 it was ligand affinity chromatography that brought to an end some 30 years of search for the
ligand-binding chain of the cell-surface Type I interferon receptor,
later named IFNAR2 (6). In 1998, using this method, we added
IL-18BP (22) to the very small family of binding proteins, proteins
which deviate from the classical definition of soluble receptors.
IL-18BP passed successfully phase I clinical studies and is a candidate for phase II in diseases where IL-18 was found deleterious. In
2006, we discovered IL-32 binding protein by pure serendipity
(23) since it was the unknown IL-32 receptor that we were trying
to find. IL-32BP is not a classical soluble receptor but rather a
protein with a dual function, an enzyme and an independent
cytokine-binding-protein. It is also a known autoantigen in autoimmune diseases. The discovery of soluble receptors and binding
proteins triggered extensive studies of their role in health and disease. The bonus of ligand affinity chromatography was not only
the ease of cloning of the corresponding gene (prior to genome
212
project era) but mainly the concomitant revelation of function of

the newly isolated protein.
The concept of proteome was introduced more than 10
years ago, followed by large-scale studies of protein expression,
localization, activities, and interactions, leading to extensive
research and technology development. Proteomics is expansively
applied in many areas, ranging from basic research, various disease
and tumor diagnoses and biomarker discovery, to therapeutic
applications. Several proteomics approaches have been developed
for protein separation and identification and for the characterization
of protein function and structure. Affinity chromatography and
mass spectrometry are among the indispensable methods that
enabled this meteoric progress (4345).
5. Notes
1. If possible, check by a bioassay that the ligand to be covalently
bound to the resin for ligand affinity column is active.
2. Use excess of ligand over target protein to be purified in a
minimal resin volume, in order to maximize yield and minimize nonspecific binding.
3. Use sufficient amounts of a receptor source to ensure a reasonable final yield taking into account losses during the purification
steps. Note that the level of most soluble cytokine receptors
found so far in normal body fluids ranges from 0.5 to 5 ng/mL.
If a cell surface receptor is to be purified, try to calculate
the minimum amount of cells or tissue needed, based on the
calculated number of these receptors per cell (e.g., deduced
from binding studies with a labeled ligand) and their apparent
molecular weight (deduced from cross-linking studies with a
labeled ligand).
4. Use polypropylene or polyethylene (opaque) rather than polystyrene (transparent) tubes throughout the study in order to
minimize protein losses by adsorption to surfaces. Also avoid
unnecessary dilution of the protein solution and keep stocks of
proteins at a high concentration.
5. Always spin (10,000 g, 15 min, 4C) the Load fraction prior
to loading on the affinity column to avoid blocking of the column. A blocked column can be re-suspended. If flow is still
blocked transfer the resin to a new cartridge or use the resin in
a batch adsorption and elution manner.
6. Some analyses (e.g., SDS-PAGE, 125I-labeling) require pure
proteins. When protein purity is not critical (e.g., bioassay)
dilute the pure protein in diluents containing a carrier protein,
e.g., 0.11% (w/v) bovine serum albumin or culture medium
containing 110% (w/v) fetal bovine serum.
12
213
References
1. Cuatrecasas, P., Wilchek, M., Anfinsen, C.B.
(1968) Selective enzyme purification by affinity
chromatography. Proc. Natl. Acad. Sci. USA
61: 636643.
2. Uhlen, M. (2008) Affinity as a tool in life science. BioTechniques 44: 649654.
3. Roque, A.C., Lowe, C.R. (2008) Affinity chromatography: history, perspectives, limitations
and prospects. Methods Mol. Biol. 421: 121.
4. Tartaglia, L.A., Goeddel, D.V. (1992) Two
TNF receptors. Immunol. Today 13: 151153.
5. Chizzonite, R., Truitt, T., Kilian, P.L., Stern,
A.S., Nunes, P., Parker, K.P., Kaffka, K.L.,
Chua, A.O., Lugg, D.K., Gubler, U. (1989)
Two high-affinity interleukin 1 receptors represent separate gene products. Proc. Natl. Acad.
Sci. USA 86: 80298033.
6. Novick, D., Cohen, B., Rubinstein, M. (1994)
The human interferon alpha/beta receptor:
characterization and molecular cloning. Cell
77: 391400.
7. Domanski, P., Witte, M., Kellum, M., Rubinstein,
M., Hackett, R., Pitha, P., Colamonici, O.R.
(1995) Cloning and expression of a long form
of the beta subunit of the interferon alpha beta
receptor that is required for signaling. J. Biol.
Chem. 270: 2160621611.
8. Marcon, L., Fritz, M.E., Kurman, C.C., Jensen,
J.C., Nelson, D.L. (1988) Soluble Tac peptide
is present in the urine of normal individuals
and at elevated levels in patients with adult
T cell leukaemia (ATL). Clin. Exp. Immunol.
73: 2933.
9. Levine, S.J. (2008) Molecular mechanisms of
soluble cytokine receptor generation. J. Biol.
Chem. 283: 1417714181.
10. Novick, D., Engelmann, H., Wallach, D.,
Rubinstein, M. (1989) Soluble cytokine receptors are present in normal human urine. J. Exp.
Med. 170: 14091414.
11. Engelmann, H., Novick, D., Wallach, D.
(1990) Two tumor necrosis factor-binding
proteins purified from human urine. Evidence
for immunological cross-reactivity with cell
surface tumor necrosis factor receptors. J. Biol.
Chem. 265: 15311536.
12. Abeck, D., Korting, H.C., Zaba, R., Dangor,
Y., Fehler, G., Ballard, R.C. (1990) Soluble
interleukin-2 receptors in serum and urine of
patients with chancroid and their response to
therapy. Int. J. STD AIDS 1: 282284.
13. Christie, G., Dacey, I., Weston, B.J. (1995)
Identification of a soluble, high affinity human
interleukin 4 binding protein in normal human
urine. Cytokine 7: 305310.
14. Fanslow, W.C., Clifford, K., VandenBos, T.,

Teel, A., Armitage, R.J., Beckmann, M.P. (1990)
A soluble form of the interleukin 4 receptor in
biological fluids. Cytokine 2: 398401.
15. Zhang, J.G., Hilton, D.J., Willson, T.A.,
McFarlane, C., Roberts, B.A., Moritz, R.L.,
Simpson, R.J., Alexander, W.S., Metcalf, D.,
Nicola, N.A. (1997) Identification, purification, and characterization of a soluble interleukin (IL)-13-binding protein. Evidence that it is
distinct from the cloned IL-13 receptor and
IL-4 receptor alpha-chains. J. Biol. Chem. 272:
94749480.
16. Veenbergen, S., Smeets, R.L., Bennink, M.B.,
Arntz, O.J., Joosten, L.A., van den Berg, W.B.,
van de Loo, F.A. (2010) The natural soluble form
of IL-18 receptor {beta} exacerbates collageninduced arthritis via modulation of T cell immune
responses. Ann. Rheum. Dis. 69: 276283.
17. Xu, W., Presnell, S.R., Parrish-Novak, J.,
Kindsvogel, W., Jaspers, S., Chen, Z., Dillon, S.R.,
Gao, Z., Gilbert, T., Madden, K., Schlutsmeyer,
S., Yao, L., Whitmore, T.E., Chandrasekher, Y.,
Grant, F.J., Maurer, M., Jelinek, L., Storey, H.,
Brender, T., Hammond, A., Topouzis, S., Clegg,
C.H., Foster, D.C. (2001) A soluble class II
cytokine receptor, IL-22RA2, is a naturally occurring IL-22 antagonist, Proc. Natl. Acad. Sci. U. S.
A. 98: 95119516.
18. Hayakawa, H., Hayakawa, M., Kume, A.,
Tominaga, S. (2007) Soluble ST2 blocks interleukin-33 signaling in allergic airway inflammation. J. Biol. Chem. 282: 2636926380.
19. Meissner, U., Blum, H., Schnare, M.,
Rollinghoff, M., Gessner, A. (2001) A soluble
form of the murine common gamma chain is
present at high concentrations in vivo and suppresses cytokine signaling. Blood 97: 183191.
20. Yasuda, H., Shima, N., Nakagawa, N.,
Mochizuki, S.I., Yano, K., Fujise, N., Sato, Y.,
Goto, M., Yamaguchi, K., Kuriyama, M.,
Kanno, T., Murakami, A., Tsuda, E., Morinaga,
T., Higashio, K. (1998) Identity of osteoclastogenesis inhibitory factor (OCIF) and osteoprotegerin (OPG): a mechanism by which OPG/
OCIF inhibits osteoclastogenesis in vitro.
Endocrinology 139: 13291337.
21. Elson, G.C., Graber, P., Losberger, C., Herren, S.,
Gretener, D., Menoud, L.N., Wells, T.N.,
Kosco-Vilbois, M.H., Gauchat, J.F. (1998)
Cytokine-like factor-1, a novel soluble protein,
shares homology with members of the cytokine
type I receptor family. J. Immunol. 161:
13711379.
22. Novick, D., Kim, S.H., Fantuzzi, G., Reznikov,
L.L., Dinarello, C.A., Rubinstein, M. (1999)
214
23.
24.
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.

Interleukin-18 binding protein: a novel modulator of the Th1 cytokine response. Immunity
10: 127136.
Novick, D., Rubinstein, M., Azam, T., Rabinkov,
A., Dinarello, C.A., Kim, S.H. (2006) Proteinase
3 is an IL-32 binding protein. Proc. Natl. Acad.
Sci. USA 103: 33163321.
Kim, S.H., Han, S.Y., Azam, T., Yoon, D.Y.,
Dinarello, C.A. (2005) Interleukin-32:
a cytokine and inducer of TNFalpha. Immunity
22: 131142.
Frosch, M., Foell, D. (2004) Wegener granulomatosis in childhood and adolescence. Eur. J.
Pediatr. 163: 425434.
Fischer, D.G., Tal, N., Novick, D., Barak, S.,
Rubinstein, M. (1993) An antiviral soluble
form of the LDL receptor induced by interferon. Science 262: 250253.
Engelmann, H., Aderka, D., Rubinstein, M.,
Rotman, D., Wallach, D. (1989) A tumor
necrosis factor-binding protein purified to
homogeneity from human urine protects cells
from tumor necrosis factor toxicity. J. Biol.
Chem. 264: 1197411980.
Hock, R.A., Hollenberg, M.D. (1980)
Characterization of the receptor for epidermal
growth factor-urogastrone in human placenta
membranes. J. Biol. Chem. 255: 1073110736.
Rubinstein, M. (1979) Preparative high perofmance liquid chromatography of proteins.
Marshall, T., Williams, K.M. (1984) Artifacts
associated with 2-mercaptoethanol upon high
resolution two-dimensional electrophoresis.
Oakley, B.R., Kirsch, D.R., Morris, N.R.
(1980) A simplified ultrasensitive silver stain
for detecting proteins in polyacrylamide gels.
Hunter, W. M. (1978) Radioimmunoassay. The
Hanbook of Experimental Immunology (D. M.
Weir, Ed.), Blackwell Press, Oxford, 14.11-14.30.
Bolton, A.E., Hunter, W.M. (1973) The labelling of proteins to high specific radioactivities
by conjugation to a 125I-containing acylating
agent. Biochem. J. 133: 529539.
Novick, D., Cohen, B., Rubinstein, M. (1992)
Soluble interferon-alpha receptor molecules are
present in body fluids. FEBS Lett. 314: 445448.
Mosley, B., Beckmann, M.P., March, C.J.,
Idzerda, R.L., Gimpel, S.D., VandenBos, T.,
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
Friend, D., Alpert, A., Anderson, D., Jackson, J.,

Wignall, J.M., Smith, C., Gallis, B., Sims, J.E.,
Urdal, D., Widmer, M.B., Cosman, D., Park,
L.S. (1989) The murine interleukin-4 receptor:
molecular cloning and characterization of secreted
and membrane bound forms. Cell 59: 335348.
Aderka, D., Engelmann, H., Maor, Y.,
Brakebusch, C., Wallach, D. (1992) Stabilization
of the bioactivity of tumor necrosis factor by its
soluble receptors. J. Exp. Med. 175: 323329.
Narazaki, M., Yasukawa, K., Saito, T., Ohsugi, Y.,
Fukui, H., Koishihara, Y., Yancopoulos, G.D.,
Taga, T., Kishimoto, T. (1993) Soluble forms of
the interleukin-6 signal-transducing receptor
component gp130 in human serum possessing a
potential to inhibit signals through membraneanchored gp130. Blood 82: 11201126.
Kim, S.H., Eisenstein, M., Reznikov, L.,
Fantuzzi, G., Novick, D., Rubinstein, M.,
Dinarello, C.A. (2000) Structural requirements
of six naturally occurring isoforms of the IL-18
binding protein to inhibit IL-18. Proc. Natl.
Acad. Sci. USA 97: 11901195.
Xiang, Y., Moss, B. (1999) Identification of
human and mouse homologs of the MC51L53L-54L family of secreted glycoproteins
encoded by the Molluscum contagiosum poxvirus. Virology 257: 297302.
Alcami, A. (2003) Viral mimicry of cytokines,
chemokines and their receptors. Nat. Rev.
Immunol. 3: 3650.
Cohen, B., Novick, D., Barak, S., Rubinstein,
M. (1995) Ligand-induced association of the
type I interferon receptor components. Mol.
Cell. Biol. 15: 42084214.
Novick, D., Rubinstein, M. (2007) The tale of
soluble receptors and binding proteins: from
bench to bedside. Cytokine Growth Factor Rev.
18: 525533.
Chen, C.H. (2008) Review of a current role of
mass spectrometry for proteome research.
Anal. Chim. Acta 624: 1636.
Sparbier, K., Wenzel, T., Dihazi, H., Blaschke,
S., Mller, G.A., Deelder, A., Flad, T., Kostrzewa,
M. (2009) Immuno-MALDI-TOF MS: New
perspectives for clinical applications of mass
spectrometry. Proteomics 9: 14421450.
Conrotto P, Souchelnytskyi S. (2008)
Proteomic approaches in biological and medical sciences: principles and applications. Exp.
Oncol. 30: 17180.
Chapter 13
In Vitro Stimulation and Detection of IFNa Production
in Human Plasmacytoid Dendritic Cells
William C. Adams and Karin Lor
Abstract
Type-1 interferons (IFNs), including IFN/, are a family of cytokines produced rapidly upon pathogen
encounter and crucial for bridging innate and adaptive immunity. IFN has been widely appreciated as a
multifunctional cytokine involved particularly in early immune responses against viral, bacterial, and parasitic infections. Although most cells may be competent to produce IFN during specific conditions, plasmacytoid dendritic cells (PDCs) are unique in their capacity to produce rapid and robust levels in response
to various pathogens. PDCs to a great extent utilize toll-like receptor (TLR) 7 and 9, localized in early
endosomes, to sense pathogen-associated nucleic acids, and initiate the signaling cascade leading to induction of IFN. Here, we provide basic protocols for the detection of IFN in individual immune cells,
particularly PDCs, using flow cytometry. We discuss the key elements for successful isolation of PDCs,
stimulation, immunostaining, and identification of IFN producing cells.
Key words: IFNalpha, Cytokine, Flow cytometry, Plasmacytoid dendritic cells, Toll-like receptor,
Intracellular cytokine staining
1. Introduction
Interferons (IFNs) were first described as antiviral agents nearly 50
years ago, and since then the field has made significant progress in
further elucidating their role in immunity. Type-1 IFNs, a family of
innate cytokines consisting of multiple isoforms of IFN and a
single IFN, have proven not only antiviral activities but also
recently been attributed important roles in bacterial and parasitic
infections. A unifying mechanism for type-1 IFN production has
been established upon early innate immune detection after the recognition of pathogen-associated molecular patterns (PAMPs) by
the infected host. Receptors that play an essential role in detecting
215
216
W.C. Adams and K. Lor
pathogen-associated nucleic acids such as, double-stranded RNA,

single-stranded RNA, and unmethylated CpG-containing DNA
include retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs)
and Toll-like receptors (TLRs) (1, 2). Engagement of such receptors by their cognate ligands initiates MyD88-independent and
-dependent signaling cascades, respectively, leading toward the
production of type-1 IFN. Most cells in the body, including nonimmune cells, are equipped with RLRs and are capable of producing type-1 IFNs. However, plasmacytoid dendritic cells (PDCs)
were recently identified as the hematopoietic immune cell population having a specialized and unique ability to rapidly produce
large amounts of type-1 IFN in response to viral infections (3, 4).
PDCs comprise a distinct subset of lymphoid-origin DCs and usually represent less than 0.5% of peripheral blood mononuclear cells
in humans. A emerging body of literature indicates that PDCs play
a unique and crucial role in providing host immunity toward a
variety of diseases. Multiple regulatory mechanisms involving
surface receptors, intracellular and exogenous factors as well as
virally encoded molecules have been shown to modulate the type-1
IFN responses in these cells. PDCs express early endosomal TLR7
and 9, which are utilized for detecting pathogen-associated ssRNA
and unmethylated CpG oligodeoxynucleotides (CpG ODN),
respectively.
Protocols for induction of IFN by various single TLR-ligands
and viruses, as well as detection of IFN, have been developed in
our laboratory. In this chapter, we mainly focus on the procedure
for intracellular cytokine staining (ICS) at the protein level in individual cells using flow cytometry. The major challenges in these
procedures lie in the critical steps of determining the production
kinetics, identifying, and handling rare delicate PDCs, as well as
using a correct mix of cell fixation and permeabilization reagents in
combination with suitable IFN-specific antibodies. For specific
antibodies to have access to and bind intracellular cytokines, the
outer cell membrane and Golgi-endoplasmic reticulum complex
membrane must be permeabilized. Prior fixing of the cells helps
the cells to withstand the effects of the permeabilizing reagents. To
this end, an optimal combination of fixative and permeabilizing
reagents will help preserve cellular morphology and protein antigenicity, minimize cell aggregation, maximize yield, and allow for
detection of the highest frequency of cytokine producing cells. For
detection of IFN, we have achieved the best results by combining
phosphate-buffered formaldehyde fixation followed by permeabilization of the cellular membranes with saponin. Cross-linking aldehyde fixatives, including formaldehyde, paraformaldehyde, and
glutaraldehyde form methylene bridges between amino, guanidyl,
hydroxyl, carboxyl groups of proteins. In this particular protocol,
we have found that formaldehyde is superior to these fixation
reagents, and that saponin is a superior permeabilizing reagent
13
In Vitro Stimulation and Detection of IFN Production...
217
compared to Tween-20, acetone, ethanol, methanol, Triton-X, or

n-octyl--D-glucopyranoside. Saponin, a plant glycoside with high
affinity for cholesterol, forms pores in cellular membranes by eluting cholesterol. Because the reaction is reversible, saponin must
remain present during all incubations and washes in the staining
procedure. The advantage to using saponin is that due to its specific interaction with cholesterol, the membrane structure and
morphology of the formaldehyde treated cells remains preserved.
However, a disadvantage of the outlined protocols is that they cannot be used for intranuclear staining as saponin does not sufficiently permeabilize the nuclear membrane.
2. Materials
2.1. Isolation
of Human
Plasmacytoid
Dendritic Cells
2.1.1. PDC Sorting
Reagents
1. CD304 (BDCA-4/Neuropilin-1) microbead kit human

(Miltenyi Biotec, Bergisch Gladbach, Germany).
2. MACS running buffer (pH 7.2) (Miltenyi Biotec). Alternatively,
a solution can be prepared containing phosphate-buffered
saline (PBS, pH 7.2) supplemented with 0.5% (v/v) bovine
serum albumin fraction V (Invitrogen, Carlsbad, CA), and
2 mM EDTA.
3. RosetteSep human monocyte enrichment cocktail (Stem Cell
Technologies, Grenoble, France).
2.1.2. PDC Sorting

Equipment
2.1.3. Cell Culture Media
2.1.4. Tubes and Filter

for Cell Culture
1. AutoMACS (Miltenyi Biotec).

2. Or MACS columns for positive selection (i.e., MS, LS, or XS)
with recommended magnet-separators and racks (Miltenyi
Biotec).
RPMI-1640 complete media supplemented with 10% (v/v) heatinactivated fetal bovine serum (FBS), 2 mM l-glutamine, 100 U/mL
penicillin, 100 M streptomycin (all from Sigma-Aldrich, St. Louis,
MO), and recombinant human IL-3 (1 ng/mL; R&D Systems,
Abingdon, England) to maintain PDC viability.
1. 5 mL polystyrene round bottom tubes.
2. 15 mL and/or 50 mL polystyrene conical tubes.
3. Sterile cell strainer with 70-m pore size (BD Biosciences).
2.2. Reagent
Preparation
2.2.1. Cell Stimulation
Examples of TLR Ligands:

1. TLR7/8: Thiazoloquinoline or various derivatives of the imidazoquinoline compound R848, and single-stranded RNAs including ssPolyU, ssRNA40, and ssRNA-DR (all from Invivogen).
218
2. TLR9: Stimulatory CpG ODN class A 2336 (CpG A), CpG

ODN class B 10103 (CpG B), and CpG ODN class C 2395
(CpG C) obtained from Coley Pharmaceutical Group (Ottawa,
Canada) or available from Invivogen (5).
Examples of Whole Virus Stocks:
1. Human Immunodeficiency Virus type-1 (HIV-1) prepared as
previously described (6, 7) (see Note 1).
2. Recombinant Adenovirus prepared as previously described (8, 9).
2.2.2. Enhancing
and Blocking TLR Signaling
and IFN Production
1. Inhibitors of endosomal acidification: Chloroquine, NH4Cl,

Bafilomycin A1 from Streptomyces griseus (all from SigmaAldrich).
2. Mouse anti-human IFN neutralizing monoclonal antibody
(mAb) (clone MMHA-11; PBL InterferonSource, Piscataway,
NJ).
3. Mouse anti-human IFN/-Receptor neutralizing mAb
(Clone MMHAR-1, PBL InterferonSource) (see Note 2).
4. Suppressive CpG ODN (TTAGGG)4 (Invivogen).
5. DOTAP cationic liposomal transfection reagent (1,2-dioleoyloxy-3-trimethylammonium-propane lipid; Roche Applied
Science, Penzberg, Germany).
2.2.3. Cytokine Protein

Transport Inhibitor
1. Brefeldin A from Penicillium brefeldianum (Sigma-Aldrich).
2.2.4. Cell Wash, Fixation,

and Permeabilization
Solutions
1. Phosphate-buffered saline (PBS, pH 7.2).
2. Monensin (e.g., Golgistop, BD Biosciences).
2. 10 Hypotonic red blood cell lysis buffer: 83 g/L NH4Cl,

10 g/L KHCO3, and 0.38 g/L EDTA in sterile deionized
water (all from Sigma-Aldrich). Dilute with water to obtain 1
working dilution.
3. Antibody staining wash solution: PBS supplemented with 2%
(v/v) FBS and 0.02% (w/v) sodium azide.
4. BD cytofix/cytoperm kit (BD Biosciences). The kit includes
both fixation and permeabilization solution (containing formaldehyde) and 10 perm/wash buffer (containing saponin).
Dilute 10 perm/wash buffer with sterile water to obtain 1
working dilution.
5. Alternative fixation solution: Formaldehyde (16%, v/v) diluted
in PBS to a final formaldehyde concentration of 4% (v/v) and
adjusted to pH 7.4. Light sensitive, store at 4C in the dark;
prepare working dilution fresh prior to fixation. Do not use
diluted solution for more than 7 days.
6. Alternative perm solution: 1 Earls buffered salt solution
(EBSS) (Gibco, Paisley, UK) with 0.1% (w/v) saponin
13
219
(Riedel-de Haen AG, Seelze, Germany) supplemented with

0.1 M HEPES Buffer (Gibco). Adjust the pH in the wash buffers to 7.27.4 with NaOH.
2.3. Intracellular IFNa
Staining
2.3.1. Fluorescent Dye
Conjugated Monoclonal
Antibodies
1. Mouse anti-human IFN mAb (clone MMHA-11; PBL

InterferonSource) (9) (see Note 3).
2. Mouse anti-human IFN-2 mAb (clone 225.C; Chromaprobe,
Maryland Heights, MO) (10).
3. Mouse anti-human IFN mAb (clone MMHB-3; PBL
InterferonSource).
4. Mouse anti-human HLA-DR (clone L243), CD3 (clone SK7),
CD11c (clone S-HCL-3), CD14 (clone MP9), CD19 (4G7),
CD56 (NCAM16.2), CD80 (clone L307.4), CD83 (clone
HB15e), CD86 (clone 2331), and CD123 (clone 9F5) mAbs
from BD Biosciences. Mouse anti-human CD123 (clone
AC145), CD303/BDCA-2 (clone AC144), and CD304/
BDCA-4 (clone AD5-17F6) mAbs from Miltenyi Biotec.
Additional clones recommended in ref. 11.
2.3.2. Conjugation
of Monoclonal Antibodies
Purified (bovine serum albumin-free) anti-IFN mAbs were conjugated to various fluorescent dyes (e.g., Alexa Fluor 647,
Invitrogen). For further information, see ref. 12.
2.3.3. Flow Cytometry

Equipment and Analysis
Software
1. Suitable flow cytometer such as BD FACS Caliburm, Canto or

LSR II (BD Biosciences).
2.3.4. Enzyme-Linked
ImmunoSorbent Assays
for Detection of IFNa
Secretion
Human multisubtype IFN ELISA kit (PBL InterferonSource)

or Human Pan-specific IFN ELISA kit (Mabtech, Nacka Strand,
Sweden).
2. Suitable analysis software such as FlowJo (Tree Star Inc.,

Ashland, OR).
3. Methods
3.1. Isolation
of Human
Plasmacytoid
Dendritic Cells
Positive Selection of PDCs

1. Obtain cell fractions enriched for monocytes and DCs (see
Note 4).
(a) Either through automated aphaeresis and subsequent
counterflow elutriation centrifugation yielding high numbers of enriched monocyte sized cells.
(b) Or, through RosetteSep enrichment of monocytes and
dendritic cells and Ficoll gradient centrifugation.
220
2. To exclude cell aggregates and debris, pass entire cell fraction

through filter into 15 or 50 mL polystyrene tube.
3. Wash cells once with PBS and centrifuge (500 g, 5 min). After
centrifugation, discard the supernatant by inverting tube.
4. Add 510 mL 1 hypotonic red blood cell lysis buffer to agitated cell pellet. Incubate for 15 min at RT (room temprature). Inactivate red blood cell lysis buffer and wash cells once
by adding 1550 mL MACS running buffer and centrifuge
(500 g, 5 min). After centrifugation, discard the supernatant
by rapidly inverting tube and then agitate pellet. Repeat step if
lysis is incomplete.
5. Add 30 L/1 108 total cells FcR blocking reagent (Human
IgG) and 75 L/1 108 total cells mouse anti-human CD304/
BDCA-4 mAb conjugated to magnetic microbeads.
7. Wash once by adding 1550 mL MACS running buffer and
centrifuge (500 g, 5 min). After centrifugation, discard the
supernatant by rapidly inverting tube.
8. Resuspend the cell pellet in MACS running buffer.
9. Start procedure for positive selection of PDCs using either
AutoMACS or appropriate columns for manual separation.
3.2. Preparation
and Stimulation
of Isolated PDCs
or Total PBMCs
Induction of IFN in sorted PDCs or unfractioned PBMC

cultures
1. Resuspend sorted PDCs in cell culture media at a density of
1 106/mL in polystyrene round bottom tubes with no less
than 2 105 cells/tube. Alternatively, resuspend 25 106
unfractioned PBMCs per milliliter of cell culture media in
polystyrene round bottom tubes (see Note 5).
2. Stimulate the cells with either TLR7/8-using imidazoquinline
or TLR9-using CpG ODNs or expose the cells to virus (usually 0.110 g/mL TLR ligands, 12 106 TCID50/mL of
replicating virus, MOI 1100 of replication-incompetent virus)
in polystyrene round bottom tubes (see Note 6).
3. If applicable, add transfectant reagents (e.g., DOTAP) together
with TLR-ligands (see Note 7).
4. If applicable, add inhibitors of endosomal acidification such as
chloroquine (550 M), Bafilomycin A1 (100500 nM), or
ammonium chloride (NH4Cl; 15 mM) (see Note 7).
5. Incubate cells at 37C and 5% CO2 for defined periods (see
Note 8).
6. Add 7 L/mL monensin (Golgistop) and 10 g/mL brefeldin
A during the final 116 h of stimulation if detection by flow
cytometry is to be performed (see Note 9).
13
3.3. Intracellular
Cytokine Staining
Detection of IFNa
Producing Cells by
Flow Cytometry
3.3.1. Surface Staining
221
1. Wash cells once by adding 23 mL antibody staining wash

solution/tube and centrifuge (560 g, 5 min). In order to
minimize cell loss avoid transferring cells to new tube. PDC
are nonadherent cells and therefore should be handled with
care so as to not disrupt the pellet before discarding the supernatant. After centrifugation, discard the supernatant by rapidly
inverting tube and then blotting on paper towel. About 50 L
of liquid will remain in the tube.
2. Make a mix of the fluorescent dye conjugated cell surface
marker specific mAbs in PBS, so that a volume of 50 L will be
added to each tube.
3. Agitate the cell pellet well and add 50 L of the antibody mixture to each sample.
4. Incubate for 20 min at room temperature in the dark. Wash
cells and discard supernatant as described in step 1.
3.3.2. Intracellular
Cytokine Staining
1. Fix the cells in 250 L Cytofix/perm buffer or 2% (v/v)

Formaldehyde for 15 min at room temperature in the dark.
2. Wash cells twice by adding 12 mL of 1 BD CytoFix wash
buffer or EBSS containing saponin, centrifuge (710 g, 5 min),
and discard the supernatant. Blot on paper towel after the
second wash.
3. Dilute fluorescent dye conjugated IFN-specific antibody in
1 BD CytoFix wash buffer, so that a volume of 50 L will be
added to each tube.
4. Add 50 L of IFN-specific antibody mix to agitated cell pellet
(see Notes 1012).
5. Incubate for 20 min at room temperature in the dark.
6. Wash cells by adding 12 mL of 1 BD CytoFix wash buffer,
centrifuge (710 g, 5 min), and discard the supernatant.
7. Agitate pellet and resuspend in 75 L of PBS.
8. Collect cells immediately by flow cytometry. Collect as many
events as possible (see Note 13).
9. Analyze flow cytometry data using appropriate software such
as Flow Jo (see Note 14).
3.4. Detection
of Secreted IFNa
Enzyme-Linked ImmunoSorbent Assay (ELISA): Alternatively,

supernatants from stimulated cells can be harvested and analyzed
for secretion of IFN by ELISA according to the manufacturers
instructions (see Note 15).
222
4. Notes
1. Extra precautions need to be taken for work with viruses.
Culture of live HIV-1 requires biosafety level L3 and Adenovirus
usually biosafety level L2. Seek advice from your institutions
biosafety officer.
2. Mouse anti-human IFN/-Receptor neutralizing mAb
(Clone MMHAR-1) strongly blocks autocrine and paracrine
signaling of IFN/ through its receptor in vitro. Blocking
can be achieved by preincubating cells with 25 g/mL of the
IFN/-Receptor mAb at 37C prior to stimulation (13).
3. Mouse anti-human IFN mAb (clone MMHA-11) binds well
to 13 of the 14 IFN species, including the dominate species
1 and 2. Additionally, this mAb is cross-reactive with Rhesus
Macaque and suitable for detection of intracellular IFN.
4. PDCs are a rare cell type usually identified based on expression
of CD123 (IL-3 receptor -chain), intermediate to high
expression of MHC class II (HLA-DR), and lack of lineage
specific markers such as CD3 (T cells), CD14 (monocytes),
CD15 (granulocytes), CD19 (B cells), and CD56 (NK cells)
(Fig. 1). In addition, PDC-specific mAbs have been developed
that provide efficient isolation of human PDC subsets (1416).
The cell surface receptors targeted by these mAbs are C-type
lectins receptors termed blood dendritic cell antigen (BDCA).
Freshly isolated PDCs express BDCA-2 (CD303) and BDCA-4
(CD304/Neuropilin-1). PDC isolation kits based on positive
selection of these markers are commercially available and offer
a convenient and rapid isolation technique of PDCs. Since
PDCs are rare cells, large blood volumes are highly recommended
Fig. 1. Expected PDC phenotype and frequency in human blood. Human peripheral blood
mononuclear cells (PBMCs) were stained with a panel of mAbs directly conjugated to fluorescent dyes and analyzed using flow cytometry. Representative contour flow plots depict
the expected phenotype (CD3/14/15/19/56, HLA-DR+, CD123+, BDCA-4+) and frequency
of PDCs in human blood.
13
223
for cell isolation. This can be achieved by utilizing automated

aphaeresis, to collect high numbers of leukocytes, and followed
subsequently with counterflow elutriation centrifugation for
enrichment of monocyte-sized cells including DCs. Sequential
magnetic separation for purification of PDCs from elutriated
monocytes, using the outlined protocol, yields high PDC
numbers (normally a range of 215 106 isolated PDCs per
10 L aphaeresis passage) and good purity (frequently >90%)
(6, 7, 9, 11, 17). The contaminating cells are mostly monocytes. Although the kits were developed for isolation of DCs
directly from unfractioned PBMCs, using PBMCs rather than
elutriated monocytes decreases the purity significantly. For that
purpose, we refined the isolation procedure of PDCs from
buffy coats by adding initial enrichment of monocytes and
DCs using RosetteSep monocyte enrichment cocktail (18).
Sorting of PDCs from buffy coats results in much lower PDC
yields (on average 1 106 isolated PDCs per buffy coat).
However, the purity is similar to that achieved using elutriated
monocytes. Moreover, it has to be taken into consideration
that BDCA-2 and BDCA-4 are also expressed on subpopulations of other cells. BDCA-4 is expressed on a subpopulation
of monocytes and is also upregulated on cultured MDCs,
which is why isolation of PDCs based on BDCA-4 expression
should only be performed on freshly isolated cells. AntiBDCA-2 binds a subpopulation of PDCs, and has been shown
to inhibit the levels of IFN production by these cells, and
thus should be avoided for this application (15).
5. The frequency of spontaneously IFN producing PDCs or
other cells within blood mononuclear cells from healthy blood
donors or even patients with ongoing illnesses is very low and
usually difficult to detect. In vivo stimulated IFN producing
cells are usually only detectable at sites of locally inflamed or
virus infected tissue. Also, IFN production in mock stimulated cell cultures is usually below detection limit and hence
the cells need in vitro stimulation to induce production and
enable detection.
6. PDCs, which express TLR7 and TLR9, respond to TLR7/
8-binding imidazoquinolines (imiquimod and R-848) and ssRNA
viruses such as HIV-1, as well as to TLR9-binding CpG ODN
(classes A and C) and DNA viruses such as Adenovirus stimulation, resulting in upregulation of IFN (6, 7, 9, 13, 1921).
In contrast, other DC subsets such as CD11c+ myeloid DCs
express a different repertoire of TLRs (i.e. TLR3, TLR4, and
TLR7/8) and respond to the corresponding TLR-ligands
accordingly. However, in contrast to TLR-ligation of PDCs,
stimulation of MDCs leads to no or very low levels of IFN
(13, 17, 21).
224
7. Acidification of endosomal compartments is required for

efficient recognition of ligands by TLR7/8 and TLR9, and for
initiation of signaling cascades leading to IFN production in
PDCs. Chloroquine, NH4Cl, and Bafilomycin A1 are welldescribed inhibitors of endosomal acidification and are therefore useful tools for studying recognition of TLR ligands. In
Fig. 3a, we show that chloroquine inhibits TLR7/8-Ligand
and Adenovirus serotype 35 induced IFN in PDCs. A range
of concentrations for each inhibitor has been provided, but it
is critical to determine the optimal concentration for each
experiment. Also, it is recommended to determine the timing
of when the inhibitors are added relative to stimulation, as they
can, for example, interfere with viral lifecycle.
DOTAP is a type of cationic liposome that forms complexes with negatively charged nucleic acids such as DNA.
Such stabilization of certain CpG ODNs, for example, leads to
retention in early endosomes and ultimately enhanced IFN
production (Fig. 3b). DOTAP and CpG ODNs may be added
simultaneously to the cells in vitro, or pre-incubated together
for a short period of time and then added to the cells. Reagents
like DOTAP may be useful experimental tools for studies of
intracellular trafficking of TLR-ligands.
8. The kinetics for IFN production in PDCs are normally very
rapid and declines rapidly thereafter. As such, initial experiments should aim to study the time-course of production for
each stimulus and/or particular cell culture system to avoid
false-negative results. We have found that PDCs have a very
rapid and strong induction of IFN in response to cognate
TLR-ligands, especially TLR7/8-Ligands. By quantitativePCR of IFN mRNA, we found that levels peaked at 4 h and
rapidly declined thereafter (13). CpG ODNs also induced
IFN mRNA transcripts albeit to much lower levels and during a prolonged time span (412 h). Furthermore, detection
of IFN protein production by ICS showed a peak at 7 h for
TLR7/8-ligand stimulation. In fact, we found that TLR7/
8-ligand induced the most rapid and robust response with high
frequencies of IFN producing PDCs at 7 h while much lower
frequencies were found in response to CpG ODNs. High levels of IFN were also found in supernatants of PDCs stimulated with any of the classes of CpG ODN or TLR7/8-Ligand.
Despite the highest detectable production of IFN by
TLR7/8-Ligand stimulation by qPCR and ICS (Fig. 2ab),
measurement of secreted IFN in the supernatants showed
that CpG ODN class C induced rather similar accumulative
secretion after 24 h (Fig. 2c). Compared to single TLR-ligands,
induction of IFN in response to wild-type, inactivated or replication-incompetent whole virus particles usually has delayed
13
225
Fig. 2. Kinetics of IFN production in PDCs. Freshly isolated PDCs were exposed to either 1 g/mL TLR7/8-Ligand or 10 g/
mL CpG ODN class C for various durations, as indicated, between 0 and 24 h in the presence or absence of brefeldin A and
monensin. (a) Representative pseudo-color flow plots of intracellular staining of IFN in PDCs after 8 h stimulation, of
which brefeldin A and monensin were present for the final 6 h. (b) PDCs were stimulated as in (a) and the duration of
monensin and brefeldin A treatment is depicted in the line graph. These cells were then fixed, permeabilized, and stained
with mAbs for IFN. The frequency (%) of IFN-producing PDCs at each time-point was determined by flow cytometry.
(c) The concentration of secreted IFN in collected supernatants was then measured by ELISA, at time points indicated in
the bar graph. The data from the two complementary assays for measuring IFN demonstrate the kinetic patterns for the
induction of IFN by two TLR ligands. TLR7/8-Ligand more rapidly induced IFN in a greater frequency of PDCs, as compared to CpG C, yet the cumulative amount of IFN in the supernatants is equivalent for both the TLR7/8-Ligand and CpG
C ODN after 24 h. Data points show mean SEM and n 3.
226
kinetics (9). This likely relates to the time required for virus
attachment, uptake or entry, and uncoating steps which are
necessary to reveal nucleic acids detectable by the cells.
9. A cytokine transport inhibitor is required for the detection of
IFN expression by flow cytometry in order to increase the
positive staining signal. Monensin and brefeldin A act to inhibit
the cellular secretory pathway, causing newly synthesized proteins to accumulate diffusely within the cytoplasm, and thus
promoting increased staining intensity readily detectable by
flow cytometry. One should be aware that monensin can also
interfere with endosomal acidification, and thus TLR signaling
(22). Therefore, it is critical to determine the time point at
which the cytokine transport inhibitors are added to cells poststimulation. Additionally, the protein transport inhibitors may
interfere with the paracrine/autocrine signaling feedback loops
required for sustained IFN production. Both monensin and
brefeldin A are toxic to the cells after longer exposure (>12 h).
Brefeldin A is soluble in DMSO, so a DMSO stimulation control is strongly recommended. Finally, it is important to consider that monensin and brefeldin A may impede mobilization
to the surface of co-stimulatory markers used for monitoring
phenotypic maturation of DCs.
10. It is possible to add both cell surface and IFN-specific antibodies simultaneously, but the system should be optimized.
The general guideline is that the more highly expressed antigens should be stained with the weaker fluorescent dyes and
vice versa. The antibodies may require different concentrations
than when they are applied separately, and some antibody
clones will not bind following fixation. Determining optimal
combinations of fluorescent dyes will also maximize separation
between populations.
11. The lower detection limit of the assay is governed by the frequency of IFN positive events in mock stimulation (e.g.,
background). Because circulating PDCs have low to no production of IFN, the background values should approach
zero. To this end, one should take the following considerations
in order to obtain optimal results: (1) titrate all antibodies
because excessively high mAb concentrations can generate
nonspecific background staining, (2) use viability dyes to
remove dead cells during the flow analysis, as mAbs bind dead
cells nonspecifically, (3) use isotype matched mAb controls,
(4) use mAbs directly conjugated to bright fluorescent dyes
such as, Alexa Fluor or Phycoerythrin (PE), and (5) remove all
unbound fluorescent dye molecules from the antibody stock.
12. In order to confirm specificity of the antibody against IFN
one can abolish the signal of the immuno-reactivity by preabsorption of the specific antibody with recombinant human
13
227
IFN. To do this, we recommend a 4C overnight incubation

of the antibody and corresponding recombinant protein at a
1:10 molar ratio and then addition of this complex to the cells
instead of primary antibody alone. Cytokine transfected cell
lines may also be used for the assessment of antibody specificity
(23, 24).
13. We stress that analysis of PDCs by flow cytometry should
immediately follow the procedure for staining of intracellular
IFN. PDC morphology is somewhat susceptible to disruption by the fixation/permeabilization treatments. Over time
the cells decrease in size such that they become difficult to differentiate from dead cells or debris.
The absolute majority of IFN producing cells should be
PDCs, so if that is not the case one has to consider problems
with staining procedure (i.e., too much mAb, unspecific binding of mAb to dead cells, autofluorescence, lack of good positive control, incorrect compensation of the flow cytometer,
etc.). Also, it is important that high event numbers are collected on the flow cytometer because PDCs are rare cells and
IFN-producing cells are normally rarer yet. For pure isolated
PDCs, this means >5 104 total events, and for whole PBMCs
>1 106 total events.
Although most cells in the body may produce IFN in
response to specific stress stimulus, PDCs are well documented
to be unique in their ability to produce very high levels of
IFN, far exceeding what has been found in any other cell
type. The detection threshold for IFN expression using ICS
and flow cytometry usually does not allow for distinct appreciation of production from other cell types than PDCs.
Therefore, it is unlikely that in a PBMC culture that significant
numbers of producer cells would be represented among other
cell types. However, confirmation of a PDC phenotype should
be performed using specific markers such as CD123 and
BDCA-4 (Fig. 1a).
14. Flow analysis is useful for determining the frequency (%) of
IFN producing PDCs. Data should preferably be presented
as pseudocolor or contour type flow plots, rather than histograms (Figs. 2 and 3). This is mainly because IFN producing
PDCs are normally a population distinct from the nonproducing cells, rather than a slight fluorescence shift, as may be
observed using histograms. In addition to using flow analysis
software to quantify the frequency of IFN producing cells, it
may also be used to determine median fluorescence intensity
(MFI). MFI is a measure of how much relative IFN is produced on a per-cell-basis. We recommend using the median
statistic as opposed to mean, because it better estimates the
central tendency of the population.
228
Fig. 3. Effect of Chloroquine and DOTAP on IFN-induction in PDCs. (a) Freshly isolated
PDCs were exposed to either TLR7/8-Ligand or recombinant Adenovirus serotype-35 for
8 h total, with brefeldin A and monensin present for the last 6 h. PDCs were also cultured
in the presence or absence of chloroquine, an inhibitor of endosomal acidification, for the
entire duration of the culture. These cells were then fixed, permeabilized, and stained with
mAbs for IFN. The frequency of IFN-producing PDCs was determined by flow cytometry. Representative flow plots shows that chloroquine treatment completely blocked IFN
production in PDCs. (b) Experiments were completed as in (a), except carried out in the
presence or absence of the cationic liposomal transfectant reagent, DOTAP, as indicated.
While DOTAP increases the frequency of PDCs which respond to TL7/8-Ligand, the
increase is more dramatic for CpG ODN class C. One representative experiment is shown
for (a) and (b).
15. It is recommended to minimize freezethaw cycles of collected

PDC supernatants in order to reduce possible IFN protein
degradation. Also, note that the lower detection threshold in
most commercially available ELISA kits is ~510 pg/mL,
which may not be sufficient for detection of IFN in most cells
13
229
or cell lines. In turn, supernatants from TLR-ligand or virus

stimulated PDCs may need to be serially diluted, because the
concentration often exceeds the upper detection limit of
commercially available ELISA kits.
5. Conclusion
We have outlined three complementary protocols in the preceding
pages, which cover; the isolation of PDCs from human blood or
identification of the cells in PBMCs, subsequent in vitro stimulation by cognate TLR-ligands or viruses, and finally detection of
IFN by ICS and analysis by flow cytometry. While we have provided information on the optimization of these protocols carried
out in our laboratory, further fine tuning on the methods may be
required depending on the particular aim of the investigation, and
the equipment, reagents, and experience in ones own laboratory.
Acknowledgments
This work was supported by grants from the Swedish Research
Council (Ventenskapsrdet), the Swedish Society for Medicine,
and the Swedish International Development Agency (SIDA).
References
1. Akira, S., Takeda, K., Kaisho, T. (2001) Tolllike receptors: critical proteins linking innate
and acquired immunity. Nat. Immunol. 2:
675680.
2. Takeuchi, O., Akira, S. (2009) Innate immunity to virus infection. Immunol. Rev. 227:
7586.
3. Cao, W., Liu, Y.J. (2007) Innate immune functions of plasmacytoid dendritic cells. Curr.
Opin. Immunol. 19: 2430.
4. Siegal, F.P., Kadowaki, N., Shodell, M.,
Fitzgerald-Bocarsly, P.A., Shah, K., Ho, S.,
Antonenko, S., Liu, Y.J. (1999) The nature of
the principal type 1 interferon-producing cells
in human blood. Science 284: 18351837.
5. Vollmer, J., Weeratna, R., Payette, P., Jurk, M.,
Schetter, C., Laucht, M., Wader, T., Tluk, S.,
Liu, M., Davis, H. L., Krieg, A.M. (2004)
Characterization of three CpG oligodeoxynucleotide classes with distinct immunostimulatory activities. Eur. J. Immunol. 34: 251262.
6. Lore, K., Smed-Sorensen, A., Vasudevan, J.,

Mascola, J.R., Koup, R.A. (2005) Myeloid and
plasmacytoid dendritic cells transfer HIV-1
preferentially to antigen-specific CD4+ T cells.
J. Exp. Med. 201: 20232033.
7. Smed-Sorensen, A., Lore, K., Vasudevan, J.,
Louder, M.K., Andersson, J., Mascola, J.R.,
Spetz, A.L., Koup, R.A. (2005) Differential
susceptibility to human immunodeficiency
virus type 1 infection of myeloid and plasmacytoid dendritic cells. J. Virol. 79: 88618869.
8. Adams, W.C., Bond, E., Havenga, M.J.,
Holterman, L., Goudsmit, J., Karlsson
Hedestam, G.B., Koup, R.A., Lore, K. (2009)
Adenovirus Serotype 5 Infects Human
Dendritic Cells via a Coxsackievirus-Adenovirus
Receptor-independent Receptor Pathway
Mediated by Lactoferrin and DC-SIGN. J.
Gen. Virol. 90: 16001610.
9. Lore, K., Adams, W.C., Havenga, M.J.,
Precopio, M.L., Holterman, L., Goudsmit, J.,
230
10.
11.
12.
13.
14.
15.
16.

Koup, R.A. (2007) Myeloid and plasmacytoid
dendritic cells are susceptible to recombinant
adenovirus vectors and stimulate polyfunctional
memory T cell responses. J. Immunol. 179:
17211729.
Ito, T., Kanzler, H., Duramad, O., Cao, W.,
Liu, Y.J. (2006) Specialization, kinetics, and
repertoire of type 1 interferon responses by
human plasmacytoid predendritic cells. Blood
107: 24232431.
Lore, K. (2004) Isolation and immunophenotyping of human and rhesus macaque dendritic
cells. Method. Cell Biol. 75: 623642.
Roederer, M. (2004). Conjugation of monoclonal antibodies: http://www.drmr.com/abcon/.
Douagi, I., Gujer, C., Sundling, C., Adams,
W.C., Smed-Sorensen, A., Seder, R.A., Karlsson
Hedestam, G.B., Lore, K. (2009) Human B
Cell Responses to TLR Ligands Are
Differentially Modulated by Myeloid and
Plasmacytoid Dendritic Cells. J. Immunol. 182:
19912001.
Dzionek, A., Fuchs, A., Schmidt, P., Cremer,
S., Zysk, M., Miltenyi, S., Buck, D.W., Schmitz,
J. (2000) BDCA-2, BDCA-3, and BDCA-4:
three markers for distinct subsets of dendritic
cells in human peripheral blood. J. Immunol.
165: 60376046.
Dzionek, A., Sohma, Y., Nagafune, J., Cella,
M., Colonna, M., Facchetti, F., Gunther, G.,
Johnston, I., Lanzavecchia, A., Nagasaka, T.,
Okada, T., Vermi, W., Winkels, G., Yamamoto,
T., Zysk, M., Yamaguchi, Y., Schmitz, J. (2001)
BDCA-2, a novel plasmacytoid dendritic cellspecific type II C-type lectin, mediates antigen
capture and is a potent inhibitor of interferon
alpha/beta induction. J. Exp. Med. 194:
18231834.
MacDonald, K.P., Munster, D.J., Clark, G.J.,
Dzionek, A., Schmitz, J., Hart, D.N. (2002)
17.
18.
19.
20.
21.
22.
23.
24.
Characterization of human blood dendritic cell

subsets. Blood 100: 45124520.
Lore, K., Betts, M.R., Brenchley, J.M., Kuruppu,
J., Khojasteh, S., Perfetto, S., Roederer, M.,
Seder, R.A., Koup, R.A. (2003) Toll-like receptor ligands modulate dendritic cells to augment
cytomegalovirus- and HIV-1-specific T cell
responses. J. Immun. 171: 43204328.
Lambert, H., Vutova, P.P., Adams, W.C., Lore,
K., Barragan, A. (2009) The Toxoplasma gondii-shuttling Function of Dendritic Cells Is
Linked to Parasite Genotype. Infect. Immun.
77: 16791688
Beignon, A.S., McKenna, K., Skoberne, M.,
Manches, O., DaSilva, I., Kavanagh, D. G.,
Larsson, M., Gorelick, R.J., Lifson, J.D.,
Bhardwaj, N. (2005) Endocytosis of HIV-1
activates plasmacytoid dendritic cells via Tolllike receptor-viral RNA interactions. J. Clin.
Invest. 115: 32653275.
Iacobelli-Martinez, M., Nemerow, G.R. (2007)
Preferential activation of Toll-like receptor nine
by CD46-utilizing adenoviruses. J. Virol. 81:
13051312.
Iwasaki, A., Medzhitov, R. (2004) Toll-like
receptor control of the adaptive immune
responses. Nat. Immunol. 5: 987995.
Mollenhauer, H.H., Morre, D.J., Rowe, L.D.
(1990) Alteration of intracellular traffic by
monensin; mechanism, specificity and relationship to toxicity. Biochim. Biophys. Acta 1031:
225246.
Andersson, J., Nagy, S., Bjork, L., Abrams, J.,
Holm, S., Andersson, U. (1992) Bacterial toxininduced cytokine production studied at the single-cell level. Immunol. Rev. 127: 6996.
Sander, B., Andersson, J., Andersson, U. (1991)
Assessment of cytokines by immunofluorescence and the paraformaldehyde- saponin procedure. Immunol. Rev. 119: 6593.
Chapter 14
A Sensitive and Versatile Cytokine Bioassay Based
on Type I Interferon Signaling in 2fTGH Cells
Lennart Zabeau, Jos Van der Heyden, and Jan Tavernier
Abstract
We have designed a sensitive and versatile bioassay for quantification of series of cytokines. The assay makes
use of chimeric receptors composed of the extracellular, ligand-binding part of the cognate cytokine receptor and the transmembrane and cytosolic part of the type I interferon receptor. Receptors can be homo(e.g. erythropoietin), di- (e.g. interleukin-5), or even trimeric (e.g. interleukin-2). Stable expression of
these chimeras in the 2fTGH cell line allows an interferon-type signaling, which makes a positive selection
in conditioned medium possible or a negative selection using a toxic guanine analog. The cytokine of
interest is quantified by the extent of cell survival or cell toxicity respectively, which can be measured by
easy and cheap crystal violet staining. This bioassay is sensitive in the lower picogram per milliliter range
and, in contrast to ELISA methods, only measures the concentration of biologically active cytokines.
Using this approach, hypersensitive 2fTGH cell lines have been developed for type I and II interferons,
erythropoietin, interleukin-2, and interleukin-5.
Key words: Bioassay, Chimeric receptors, Interferon, 2fTGH, 6-Thioguanine, Toxicity
1. Introduction
Accurate and sensitive methods for detection and quantification of
small quantities of cytokines have become crucial in clinical diagnostics. Current techniques can be roughly divided in bioassays
and immunoassays. Antibody-based techniques can allow detection of multiple cytokines simultaneously and can be adapted for
high-throughput screening. On the other hand, biological active
levels of a certain cytokine can only be measured using a bioassay.
These latter often require specific cell lines and growth media
depending on the cytokine. There is therefore a growing need for
a unified system that makes sensitive, easy, cheap, and accurate
measurements of large series of bioactive cytokines possible.
231
232
L. Zabeau et al.
The fibrosarcoma cell line 2fTGH is deficient in the gene

encoding for hypoxanthine-guanine phosphoribosyl transferase
(HGPRT), an enzyme involved in the generation of purine nucleotides through the so-called purine salvage pathway. The deficiency
can be complemented by the stable integration of the bacterial gpt
gene, coding for xanthine-guanine phosphoribosyl transferase
(XGPRT). Induced expression of this gene allows both positive
selection in hypoxanthine-aminopterine-thymidine (HAT) medium
or negative selection using the toxic guanine analog 6-thioguanine
(6-TG). A cell line with the gpt gene under control of the type I
interferon (IFN) inducible 616 promotor (1) has been proven a
very powerful tool for cloning and characterization of components
involved in IFN-/ signal transduction pathways (2, 3). This latter takes advantage of the endogenous expression of the type I IFN
receptors on the surface of 2fTGH cells.
The class II cytokine receptor IFNAR is composed of two subunits, IFNAR1 and IFNAR2-2, an IFNAR2 isoform with a complete intracellular domain (4). Like all members of the class II
cytokine receptor family, the IFNAR lacks intrinsic kinase activity
and uses the receptor-associated kinases Jak1 and Tyk2. In a generally accepted model, IFN stimulation clusters INFAR1 and
INFAR2 receptors in such a way that the kinases are brought in
close and correct proximity, allowing activation by cross-phosphorylation. Activated kinases phosphorylate tyrosine residues in the
cytoplasmic tails, thereby providing docking sites for downstream
signaling molecules, such as the signal transducers and activators of
transcription 1 (STAT1) and STAT2. Once recruited, these STATs
become a substrate of the kinase activity, and together with IRF9
(interferon regulatory factor 9) translocate to the nucleus where
they regulate gene-transcription (see Fig. 1).
In this bioassay we use chimeric receptors to combine specific
cytokine capturing by the ectodomains of their cognate receptors
with activation of the intracellular IFN signaling pathways and
IFN-induced 6-TG sensitivity in the 2fTGH cell line. In a first
step, the extracellular domains of the tested cytokine are coupled
to the transmembrane and cytoplasmic domains of IFNAR1 and
IFNAR2-2. To increase efficiency of signaling by these chimeric
receptors, additional leucine residues can be inserted in the transmembrane domain. This latter allows rotations of the cytoplasmic
tails in the activated chimeric receptor complex. Combinations of
the resulting receptors are initially tested for 616 promotor activation in transient transfection experiments, which allows selecting
the most signaling efficient combination. This combination is then
transfected in 2fTGH cells and selected for survival in HAT medium
supplemented with the tested cytokine. Clones are analyzed and
finally used to measure the cytokine-induced 6-TG toxicity.
Sensitivities of different bioassays are summarized in Table 1 (5).
STAT2
STAT1
trimeric
receptors
IFNaR1
IFNaR1
-Y P
P Y-
233
IFNaR2-2
heterodimeric
receptors
IFNaR2-2
homodimeric
receptors
IFNaR1
IFNaR1
IFNaR2-2
INF/
A Sensitive and Versatile Cytokine Bioassay Based on Type I
IFNaR2-2
14
IRF9
Survival in HAT
or
Cell death in 6-TG
E. coli gpt
6-16 promotor
Fig. 1. Type I interferon and chimeric receptors and their downstream signal transduction pathway leading to activation of
E. coli gpt transcription.
Table 1
6-TG cell survival assay
Cytokine
Type receptor
EC50
Detection limit
IFN-/
Heterodimeric
~5 pg/mL
~2 pg/mL
IFN-
Heterodimeric
~20 pg/mL
~5 pg/mL
Erythropoietin
Homodimeric
~3 pg/mL
~1 pg/mL
Interleukin-2
Heterotrimeric
~6 pg/mL
~2 pg/mL
Interleukin-5
Heterodimeric
~15 pg/mL
~4 pg/mL
2. Materials
2.1. Construction
of Chimeric Receptors
Restriction enzymes (5 U per reaction; Bioloabs) and T4 DNA

ligase (1 U per reaction; Invitrogen) are used according to the
manufacturers guidelines. Synthetic oligonucleotides are ordered
from Eurogentec.
2.2. Transient
Evaluation of Receptor
Combinations
1. Culture medium: Dulbeccos modified Eagles medium

(DMEM) supplemented with 10% fetal bovine serum and antibiotics (penicillin/streptomycin; Invitrogen). FCS is inactivated
(20 min at 56C) for 2fTGH cell culture.
234
L. Zabeau et al.
2. CaCl2 buffer: prepare a 2.5 M CaCl2 stock and filter-sterilize

(store at 20C).
3. 2 HeBS (Hepes-buffered saline) buffer: 280 mM NaCl,
1.5 mM Na2HPO4 and 50 mM Hepes in double distilled water,
adjust pH to 7.05 with 1 M NaOH (see Note 3), filter-sterilize
and store in aliquots at 20C (avoid repeated freeze-thaw
cycles).
4. SEAP activity is measured with the Phospha-Light kit (Tropix)
using the luminogenic CSPD (disodium 3-(4-methoxyspiro
(1,2-dioxetane-3, 2-(5-chloro) tricyclo[3.3.1.1]decan)-4-yl)
phenyl 1 phosphate) substrate.
5. The IFN used in this assay is human IFN-2b (Pepro Tech).
2.3. Generation
of Stable Cell Lines
2.4. Bioassay
Selection medium: HAT concentrate (containing 5 mM sodium

hypoxanthine, 20 M aminopterin, 0.8 mM thymidine; Invitrogen)
1/50 diluted in culture medium supplemented with penicillin/
streptomycin (Invitrogen).
1. Selection medium for cell survival: HAT in culture medium.
Selection medium for cell toxicity: 50 g/mL 6-TG (SigmaAldrich) in culture medium.
2. Crystal violet staining solution: 0.5% crystal violet (w/v) in 3%
formaldehyde, 30% ethanol, and 0.17% NaCl (w/v).
3. Solubilization buffer: 33% acetic acid; OD reading at 595 nm.
3. Methods
The protocol is organized in four sections, with the first three parts
being optimization, the fourth the actual bioassay. The optimization
includes the construction of chimeric receptors (see Subheading 3.1)
and evaluation of efficiency of signaling using transient transfection
experiments (see Subheading 3.2) (see Note 5). Once an optimal
receptor combination is selected, these are stably transfected in
the 2fTGH cell line (see Subheading 3.3). The last section (see
Subheading 3.4) deals with the cytokine-induced 6-TG toxicity
bioassay.
3.1. Construction
of Chimeric Receptors
In this procedure, the extracellular (EC) parts of the receptor for

the tested cytokine are fused to the transmembrane (TM) and
intracellular (IC) domains of IFNAR1 or IFNAR2-2. In case of a
homomeric receptor, this results in two chimeric receptors. When
the receptor is composed of two different chains, each receptor
subunit is fused to IFNAR1 or IFNAR2-2, leading to four different chimeras. To further enhance signaling, additional leucine residues (one, two or three) can be inserted in the TM domain of each
14
235
chimeric receptor by site-directed mutagenesis (see Note 6). In the

case of a heterodimeric receptor this involves 16 combinations to
evaluate transiently (see Note 1).
1. Amplify the EC part of the receptor(s) of interest, as well as the
TM + IC domains of IFNAR1 and INFaR2-2 using standard
PCR techniques.
2. Cut fragments and ligate in an opened pcDNA3 vector
(Invitrogen).
3. Transform in E. coli.
4. Purify plasmid using a method yielding DNA suitable for
transfection.
5. Insert one, two, or three extra leucines in the TM domain using
the QuikChange Site Directed Mutagenesis Kit (Stratagene).
3.2. Transient
Evaluation
of Receptor
Combinations
In what follows, combinations of the different chimeric receptors

are tested for efficiency of type I interferon signaling by transient
transfection experiments. A reporter construct with the secreted
alkaline phosphatase (SEAP) under control of the 616 promotor
is thereby used to quantify receptor activation (6) (see Note 5).
Day 0: Seed 400,000 2fTGH cells per 10 cm2 well.
Day 1: Transfect chimeric receptors (or a combination thereof)
together with 616 SEAP reporter (2 g DNA in total)
using standard calcium phosphate transfection techniques.
Day 2: Wash cells with PBS without calcium and magnesium.
Day 3: Trypsinize cells with 200 L trypsin (Invitrogen), seed in a
96-well plate and stimulate overnight with a serial dilution
of the cytokine or IFN- as a positive control.
Day 4: Lyse the cells in 10 L lysis buffer for 10 min and measure
alkaline phosphatase activity using 50 L of a 1/20 dilution
of the CSPD substrate in a chemiluminescence reader for
10 s (TopCount, Perkin Elmer).
3.3. Generation
of Stable Cell Lines
Next, the most efficient chimeric receptor combination is stably

expressed in 2fTGH cells and selected for cytokine-dependent
growth in HAT medium. After 2 weeks of selection, individual
clones are isolated and screened for highest sensitivity.
1. Day 1: seed 7 106 cells in a 175 cm2 flask and culture overnight.
2. Day 2: transfect 35 g chimeric receptor(s) with calcium phosphate overnight.
3. Day 3: wash cells with PBS and culture overnight.
4. Day 4: trypsinize and subculture 1/10.
5. Replace with culture medium supplemented with HAT and
suboptimal concentrations of the appropriate cytokine.
Eventually different cytokine concentrations can be applied.
236
L. Zabeau et al.
6. Replace medium every 4 days.

7. Colony picking: remove medium and cover each colony with a
small paper disk (3MM, Whatman) that has been soaked in
trypsin. After a couple of minutes wipe off the colony with the
paper disk using sterile forceps.
8. Transfer to a 24- or 48-well plate.
9. Upscale cultures.
3.4. Bioassay
The expanded clonal cell lines are finally screened for highest sensitivity in the cytokine-dependent 6-TG bioassay. The read-out is
the cheap and easy crystal violet staining method for living cells
quantified using a colorimeter (see Note 4).
1. Seed 3,000 to 5,000 cells per 96-well in medium containing a
serial dilution of cytokine.
2. After 24 h, add 6-TG to a final concentration of 50 g/mL.
3. Further incubate for 4 days.
4. Remove medium and stain for 10 min with 50 L crystal violet
staining solution.
5. Wash cells gently but extensively with tap water.
6. Add 100 L solubilization buffer and measure in a colorimeter
at 595 nm.
4. Notes
1. This protocol can also be adapted for trimeric receptors. In this
case, the cytokine-capturing receptor subunit is first stably
expressed in 2fTGH cells. IFNAR1 and IFNAR2-2 chimeras
of the remaining two components are then transfected in these
cells. Transient evaluation, selection, and the bioassay are
essentially the same as for homo- and dimeric receptors. Using
a 2fTGH cell line stably expressing the interleukin-2 receptor
chain, we have been able to set up a bioassay for this cytokine
(see Table 1).
2. Since type I IFN receptors are endogenously expressed on the
surface of the 2fTGH cells, a parallel assay using neutralizing
anti-IFNAR2 antibody should be used to check the presence
of type I IFNs when biological samples (like blood, urine,
sputum, ) are tested.
3. An exact pH of the 2 HeBS buffer is critical for transfection
efficiency. The optimal range is very narrow: from 7.05 to
7.12.
14
237
4. As an alternative for the crystal violet staining, the use of

ATPlite (Perkin Elmer) can be considered.
5. For transient evaluation Hek293 or Hek293T can be used
because of their high transfection efficiency resulting in a
robust luminescence signal.
6. Position of the extra leucine residues in the transmembrane
region of the IFN receptors. As an example a Pac1 restriction
site is added at the N-terminal site for the fusion with the
extracellular part of the cytokine receptor of interest.
IFNAR1 TM
Pac1
LLL
TTAATTAAAATTTGGCTTATAGTTGGAATT---------TGT
ATTGCATTATTTGCTCTCCCGTTTGTC
TTAATTAAAATTTGGCTTATAGTTGGAATTCTC------TG
TATTGCATTATTTGCTCTCCCGTTTGTC
TTAATTAAAATTTGGCTTATAGTTGGAATTCTCCTC--TGTATTGCATTATTTGCTCTCCCGTTTGTC
TTAATTAAAATTTGGCTTATAGTTGGAATTCTCCTCC
TCTGTATTGCATTATTTGCTCTCCCGTTTGTC
IFNAR2 TM
Pac1
LLL
TTAATTAAAATTTGGCTTATAGTTGGAATT---------TGT
ATTGCATTATTTGCTCTC
TTAATTAAAATTTGGCTTATAGTTGGAATTCTC------TG
TATTGCATTATTTGCTCTC
TTAATTAAAATTTGGCTTATAGTTGGAATTCTCCTC--TGTATTGCATTATTTGCTCTC
TTAATTAAAATTTGGCTTATAGTTGGAATTCTCCTCCT
CTGTATTGCATTATTTGCTCTC
References
1. Pellegrini, S., John, J., Shearer, M., Kerr, I.M.,
Stark, G.R. (1989) Use of a selectable marker
regulated by alpha interferon to obtain mutations in the signaling pathway. Mol. Cell. Biol. 9:
46054612.
2. Velazquez, L., Fellous, M., Stark, G.R., Pellegrini, S.
(1992) A protein tyrosine kinase in the interferon
alpha/beta signaling pathway. Cell 70: 313322.
3. Darnell, J.E., Jr., Kerr, I.M., Stark, G.R. (1994)
Jak-STAT pathways and transcriptional activation
in response to IFNs and other extracellular signaling proteins. Science 264: 14151421.
4. Novick, D., Cohen, B., Rubinstein, M. (1994)
The human interferon alpha/beta receptor:
characterization and molecular cloning. Cell 77:

391400.
5. Van Ostade, X., Schauvliege, L., Pattyn, E.,
Verhee, A., Vandekerckhove, J., Tavernier, J.
(2000) A sensitive and versatile bioassay for
ligands that signal through receptor clustering. J.
Interferon Cytokine Res. 20: 7987.
6. Pattyn, E., Van Ostade, X., Schauvliege, L.,
Verhee, A., Kalai, M., Vandekerckhove, J.,
Tavernier, J. (1999) Dimerization of the interferon type I receptor IFNaR2-2 is sufficient for
induction of interferon effector genes but not for
full antiviral activity. J. Biol. Chem. 274:
3483834845.
INDEX
A
Actinomycin D .................................................73, 76, 78, 87

Affinity chromatography ..............5, 7576, 8586, 195212
Affymetrix ................................................... 2552, 151154
AGO protein ........................................................... 134, 135
Antiviral activity .............................................. 25, 209, 215
Apoptosis..............................................5051, 159, 180, 189
ARE. See AU-rich element
AU-rich element (ARE) .................................72, 76, 9195,
97, 98, 100, 102
Autoimmune disease ....................................................... 197
Electrophoretic mobility shift assay (EMSA)............. 7475,

8183, 87
EMSA. See Electrophoretic mobility shift assay
End-labeling of synthetic RNA............................. 75, 8384
Epithelial cells and tissues ........................................... 5569
Epstein-Barr virus ............................................................. 25
B
Basic local alignment search tool
(BLAST) ............................................. 168, 183, 210
Bioassay .................................... 16, 201, 209, 212, 231237
Biomarkers ...................................................................... 212
BLAST. See Basic local alignment search tool
C
CD4........................................ 2631, 38, 39, 46, 52, 77, 114
Chicken -actin promoter ............................................... 125
Chimeric receptors .................................................. 232235
Chloramine-T.......................................................... 200, 206
Concanavalin A ............................................................... 3, 4
Ct-value ........................................................8, 1520, 22, 66
Cytoplasmic extracts ...............................74, 8184, 126, 127
D
Database for annotation, visualization, and integrated
discovery (DAVID) ......................................... 2552
DAVID. See Database for annotation, visualization, and
integrated discovery
DEPC. See Diethylpyrocarbonate
Diabetes........................................................................... 179
Dicer..... .............................................................134137, 180
Diethylpyrocarbonate (DEPC) ........................... 76, 86, 125
Doxycycline ........................................................... 73, 76, 79
Drosha. .................................................................... 135, 136
dsRNA-dependent protein kinase .......................... 134, 136,
137, 139, 152, 155, 164
G
Geneticin ......................................................... 119, 120, 128
GFP probes ................................................................. 74, 81
-Globin probes .......................................................... 74, 81
H
HA. See Hyaluronan
Hammerhead ribozyme ........................................... 117131
HGPRT. See Hypoxanthine-guanine phosphoribosyl
transferase
Hierarchical clustering ...................................................... 38
HIV. See Human immunodeficiency virus
House keeping gene .................................128130, 139, 190
Human antigen R (HuR) ............................................ 72, 94
Human immunodeficiency virus (HIV) ............... 2628, 51,
106, 218, 222, 223
HuR. See Human antigen R
Hyaluronan (HA) .................................................... 107109
Hypoxanthine-guanine phosphoribosyl transferase
(HGPRT) ............................................................ 232
I
ICS. See Intracellular cytokine staining
IFN-.................................. 3032, 4445, 49, 51, 148150,
196, 199, 211, 215229, 232, 233, 235
IFNAR2 ....................196, 199, 210, 211, 232, 234, 236, 237
IL32 binding protein ..............................196197, 199, 211
Inducible nitric oxide synthase ........................................ 181
Inflammation .......................... 23, 5, 72, 106, 164, 165, 174
Insulinoma....................................................................... 184
Insulitis ............................................................................ 179
Integrin.................................................................... 105115
Interferon-....................................................................... 93
DOI 10.1007/978-1-61779-439-1, Springer Science+Business Media, LLC 2012
239
CYTOKINE PROTOCOLS
240 Index
Interferon- ....................................................................... 93
Interferon-stimulated genes (ISGs) ........................ 136, 138,
148, 150, 153, 155, 163174
Interleukin1 ............................................................... 2, 56
Interleukin27 ....................................................... 2552, 93
Intracellular cytokine staining (ICS) ...................... 216, 221,
224, 227, 229
ISGs. See Interferon-stimulated genes
K
Key pathways ..................................................................... 51
Knockout model ...................................................... 189190
L
Lentivirus ................................................................ 164, 169
Leukocyte .............................................3, 106, 108, 111, 223
Ligand affinity chromatography .............................. 195212
Lipopolysaccharide (LPS) ............................................... 102
Liposome .................................. 106, 108110, 113, 114, 224
LNA. See Locked nucleic acids
Locked nucleic acids (LNA) ............................................. 14
LPS. See Lipopolysaccharide
M
Macrophage .......................... 2627, 2932, 38, 39, 165, 169
MACS. See Magnetic-activated cell sorting
Magnetic-activated cell sorting (MACS) ................... 27, 29,
30, 217, 220
MCF. See Monocytic cell factor
Melting curve analysis ......................................8, 14, 19, 172
Messenger RNA (mRNA)
decay .......................................................... 7274, 7681
knockdown ........................................ 119120, 125129
stability ......................................... 56, 91, 94, 95, 97, 103
MGB probes.................................................................... 78
Microarray .............................................2552, 59, 137139,
148155, 159, 189191
MicroRNA (miRNA)............. 5569, 91, 133141, 167, 183
miRBase ...................................................................... 58, 69
miRNA. See MicroRNA
Mitogen ..................................................................... 2, 3, 95
Molecular beacons ........................................................... 78
Monocytic cell factor (MCF) ..............................................5
mRNA. See Messenger RNA
N
NanoDrop .......................................... 10, 11, 13, 35, 97, 100
Nanoparticle ............................................................ 105115
Northern blotting .................................................. 74, 7981
O
2,5-Oligoadenylate ................................................ 137, 164
Oligofectamine .........................................137, 141, 146147
P
Pancreatic islets................................................................ 181
Pfaffl method............................................................... 1719
PKR. See Protein kinase R
Plasmacytoid dendritic cells .................................... 215229
Polyadenylation ............................................................... 102
Polyinosinic:polycytidylic acid (PolyI:C) ......................... 156
Poly-rI:rC ............................................................................3
Posttranscriptional regulation ................................ 7188, 95
Primer efficiency................................................................ 17
Protein kinase R (PKR) .................... 134, 137, 139, 152, 164
R
Real-time-quantitative PCR (RT-qPCR) ......723, 125, 189
Relative expression software tool (REST) ................... 1819
REST. See Relative expression software tool
RISC complex ............................ 69, 135, 136, 142, 180, 183
RNA-binding protein........72, 7576, 82, 8586, 94, 95, 164
RNAi. See RNA interference
RNA interference (RNAi) ................. 56, 105, 106, 133160,
163165, 168, 169, 179193
RNase L .................................................................. 137, 164
RT-qPCR. See Real-time-quantitative PCR
S
Short hairpin ................................................... 135, 137, 164
Silicic acid........................................................................ 3, 4
siRNA. See Small interfering RNA
Small interfering RNA (siRNA) .............105115, 134144,
146147, 156159, 164165, 167, 180181,
183192
Spectrometry ................................ 82, 86, 197, 209, 210, 212
Supershift assay ................................................................. 83
Surface plasmon resonance ...................................... 197, 208
Sybr-Green ................................8, 9, 14, 22, 57, 68, 166, 172
T
TaqMan ................................ 78, 5665, 119, 125127, 150
TaqMan low-density array (TLDA)........................... 56, 57,
59, 60, 6265
T cell 2526, 72, 73, 77, 78, 81, 82, 85, 86, 108
TGF- ......................................... 56, 94, 120, 121, 128131
TLDA. See TaqMan low-density array
TLR. See Toll-like receptor
Toll-like receptor (TLR) .........................138, 164, 216218,
220, 223226, 228229
Toxicity bioassay .............................................................. 234
Transfection ......................................... 76, 97, 106, 118, 135,
168, 180, 218, 232
Tristetraprolin (TTP) ............................................ 72, 9495
TTP. See Tristetraprolin
Turbofect .................................. 120, 129131, 141, 147148
Type I interferon ...................... 196, 210, 211, 215, 231237
CYTOKINE PROTOCOLS
241
Index
U
Urinary proteins .......................................198, 199, 201, 202

UV crosslinking .................................................... 75, 8082,
84, 85, 87
Xanthine-guanine phosphoribosyl transferase

(XGPRT) ............................................................ 232
XGPRT. See Xanthine-guanine phosphoribosyl transferase
Wegeners auto-antigen ................................................... 197
Zeta potential .................................................. 106, 107, 110

Cytokine Protocols - MArc Ley

Caricato da

Informazioni sul documento

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Cytokine Protocols - MArc Ley

Caricato da

Copyright:

Formati disponibili

METHODS

For further volumes:

1 A Tale of Two Interferon Bioassays: How Frustration with Discrepant Results

10 shRNA-Induced Interferon-Stimulated Gene Analysis. . . . . . . . . . . . . . . . . . . . . . . 163

KARIN LOR Department of Medicine, Center for Infectious Medicine,

1. Die Methode ist

This is exactly what happened to me and my coworkers (including

A Tale of Two Interferon Bioassays

responses to infections and tumors. In the 1970s, our laboratory

human IFNs, to set up a routine of semi-mass production of IFN-

A Tale of Two Interferon Bioassays

presence of antibodies against IFN-. We first speculated that 22 K

examples of a cytokine cascade. Induction of one cytokine by

production by a 22 K protein from human

TaqMan probes, Molecular Beacons, MGB probes, and others

we focus on the details of RNA isolation and cDNA synthesis methods,

Absolute quantification analysis ideally determines the absolute

Relative quantification analysis determines the levels of expression

1.3. Real-Time PCR

optical signals detected from each sample. Both aspects greatly

1. RNA isolation, including on-column DNase treatment: RNeasy

1. Luria Bertani (LB) medium (1 L).

Isolation and quantification of good-quality RNA (see Note 1) are

3.2. cDNA Synthesis

The calibration curves used in absolute quantification can be based

1. Prepare LB agar plates containing ampicillin, X-Gal, and

9. Make glycerol stocks by combining the remaining 1-mL overnight

Once the size of the plasmid containing the GOI is known, it is

Ex. 3,927,000 (g/mol)/6.02214199 1023 = 6.52 1018 g/molecule.

g/L)/(6.52 1018g/molecule) = 4.6 1010

Once the number of molecules in 1 L of linearized plasmid

The design of specific primers that work at a good efficiency is of

A typical PCR profile includes an initial denaturation step of

Select a cDNA template or recDNA containing the sequence of the

tubes and add 1 L of each primer to give final concentrations as

We usually find 300 nM the optimal concentration for both

Depending on the subsequent method of analysis, there are several

one or more RGs. To date, several mathematical models have been

The Ct method (9) is based on the assumption that the primers

Set the threshold to 0.1 for all genes.

Export the Ct values to Excel.

Select the sample at time point 0 (zero) as the calibrator (the

Apply the following formula:

This method has the advantage that standard curves are

This method does not require the amplification efficiency of different

Export the Ct values to Excel.

Select the sample at time point 0 (zero) as calibrator and apply

The Pfaffl method is a modification of the Ct method with

quantification software called Relative Expression Software Tool

To date, several methods have been developed to calculate the

The Ct values and the amplification efficiency for each sample

The average amplification efficiency (EA) for each primer in

Absolute quantification refers to an analysis, where unknown samples

Standard curves for several GOI can be amplified in the first

Alternatively, it is possible to export data to excel and perform

Express data as GOI (copy number)/x ng total RNA.

To normalize data and correct for variations in template input,

When optimized, standard curves are highly reproducible and

Depending on the applications, the use of technical and biological

synthesis are performed independently but under identical conditions.

Furthermore, the longer templates derived from recDNA and

After having set the threshold (0.1), export the Ct values

Calculate the LOG10 of the given arbitrary concentration