Monitoring innate immune recruitment by siRNAs in mammalian cells.

Chapter 2
Monitoring Innate Immune Recruitment by siRNAs
in Mammalian Cells
Michael P. Gantier and Bryan R.G. Williams
Abstract
The use of small interfering RNAs (siRNAs) in human therapy may be hindered by the recruitment of
nonspecific effects such as the activation of innate immune responses. Recently, several innate immune
receptors have been implicated in the detection of siRNAs. This chapter provides a brief overview of the
current knowledge of siRNA-induced innate immunity, as well as protocols for the rapid identification of
siRNAs with innate immune stimulatory activity.
Key words: Innate immunity, RNA interference, siRNA, RIG-I, TLR7, TLR8
During viral infection, mammals rely on an early detection of
foreign ribonucleic acids to mount a rapid antiviral response.
While this phenomenon has been known for more than four
decades, insights into the molecular identity of components of
the response have been gained only recently (1). Two detection
pathways have been identified in blood immune cells as directly
involved in innate immune activation by exogenous RNAs. The
cells orchestrating the initiation of this antiviral response sense
viral RNAs through Toll-like receptors (TLRs) or retinoic acid
inducible gene I (RIG-I)-like receptors (1).
Originally thought to be too small to be recognized by the
sensors of the innate immune system, small interfering RNA
(siRNA) activation of a strong innate immune response is now
well established (2). To date, four main characteristics of
siRNAs have been associated with the recruitment of innate
1. Introduction
Wei-Ping Min and Thomas Ichim (eds.), RNA Interference, Methods in Molecular Biology, vol. 623,
DOI 10.1007/978-1-60761-588-0_2, © Humana Press, a part of Springer Science + Business Media, LLC 2010
21

Gantier and Williams
immunity and subsequent cytokine production: a) Secondary
structure, which is detected by TLR3; b) uridine content, detected
by TLR 7/8; c) end terminal structure of blunt-end siRNA from
21-27 nt detected by RIG-1; and 25 nt duplexes bearing a 5’ or
3’ monophosphate, also detected by RIG-1 (3–11).
We have established different protocols that allow for rapid
discrimination among different siRNAs for their capacity to
recruit TLR7/8 and RIG-I (12, 13). Whether or not the ability
of an siRNA to induce immunostimulation through these recep-
tors is the desired outcome (14), these systems are a useful start-
ing point prior to further validation in peripheral blood
mononuclear cells (PBMCs) from animal models.
In this chapter, we describe two protocols allowing for the
evaluation of mouse TLR7 (and per se, also human TLR7) and
human TLR8 recruitment by siRNAs. We also describe a simple
real-time Reverse Transcription-Polymerase Chain Reaction
(RT-PCR) protocol, based on human T98G cells (adapted from
Marques et al. (8)).
1. RAW 264.7: ATCC reference TIB-71. T98G cells: ATCC
reference CRL-1690.
2. Ficoll-Paque Plus (GE Healthcare)
3. Lithium-heparin sterile tubes (Sarstedt, Nümbrecht,
Germany).
4. Dulbecco’s Modified Eagle’s Medium (DMEM) (Invitrogen
Corporation) supplemented with 10% sterile fetal bovine
serum (FBS; ICPBio Ltd, Auckland, New Zealand) and 1×
antibiotic/antimycotic (Invitrogen Corporation) (referred to
as complete DMEM medium).
5. Roswell Park Memorial Institute medium (RPMI) 1640 plus
l-glutamine medium (Invitrogen Corporation) comple-
mented with 1× antibiotic/antimycotic and 10% FBS (referred
to as complete RPMI 1640).
6. Dulbecco’s Phosphate-Buffered Saline (PBS, Invitrogen
Corporation).
7. TrypLE™ Express Stable Trypsin (Invitrogen Corporation).
8. Sterile tissue culture-treated microtest™ 96-well plates
(Falcon)
9. Sterile, tissue culture-treated 48-well plates (JET BIOFIL,
Guangzhou, China).
2. Materials
2.1. Cell Culture
22

Monitoring Innate Immune Recruitment by siRNAs in Mammalian Cells
10. Human TLR8 and mouse TLR7 agonist: CL75 (Invivogen,
San Diego, USA).
11. N-[1-(2,3-Dioleoyloxy)propyl]-N,N,N-trimethylammo-
nium methylsulfate (DOTAP) (Roche).
12. Opti-MEM® (Invitrogen Corporation).
13. Lipofectamine 2000 (Invitrogen Corporation).
14. siRNAs: synthesized by Integrated DNA Technologies
(IDT) as single-stranded RNAs; resuspended in filter-steril-
ized duplex buffer (100 mM potassium acetate, 30 mM
HEPES, pH 7.5) in UltraPure™ DNase/RNase-Free
Distilled Water (referred to as RNase-free H
2
O, Invitrogen
Corporation) to a concentration of 80 mM. Each duplex is
annealed at 92°C for 2 min and left for 30 min at room
temperature before being aliquoted and frozen at −80°C.
siControl is a nontargeting 21 nucleotide siRNA (siControl
1, Ambion).
1. OptEIA ELISA sets (BD Biosciences).
2. PBS 10×: NaCl 8% (w/v), KCl 0.2% (w/v), Na
2
HPO
4
1.22%
(w/v), KH
2
PO
4
0.2% (w/v) in ddH
2
O – pH 7.4 (all reagents
are from Sigma-Aldrich).
3. PBS-tween (PBST): 1× PBS diluted in H
2
O complemented
with 0.05% tween 20 (Sigma-Aldrich).
4. Pharmingen Assay Diluent (BD Biosciences Pharmingen).
5. F96 maxisorp plates (nunc, Roskilde, Denmark).
6. Tetramethyl benzidine substrate (TMB, Sigma-Aldrich)
7. Sulfuric acid 2 N (Sigma-Aldrich).
8. Plate reader with 450 nm absorbance filter.
1. NucleoSpin RNA II kit (MACHEREY-NAGEL, Düren,
Germany). Supplement RA1 buffer with 1% v/v 2-mercapto-
ethanol (Bme) (Sigma-Aldrich) immediately before adding to
the cells.
2. Superscript III Reverse Transcriptase – includes 5× first strand
buffer and 0.1 M dithiothreitol (DTT), 10 mM deoxy-nucle-
otides triphosphate (dNTPs), Oligo(dT)
20
Primer, and
RNaseOUT™ (all from Invitrogen Corporation).
3. SYBR GreenER™ qPCR SuperMix for iCycler® instrument
(Invitrogen Corporation).
4. IQ5 Multicolor Biorad i-cycler.
5. Optical Tape (Bio-Rad).
6. Multiplate 96-well clear (Bio-Rad).
2.2. Tumor Necrosis
Factor a (TNF-a)
Enzyme-Linked
ImmunoSorbent Assay
(ELISA)
2.3. RNA Extraction/
cDNA Synthesis/Real
Time
23

Gantier and Williams
First and foremost, the ability of siRNAs to recruit the innate
immune system is highly dependent on the cell type considered.
Plasmacytoid dendritic cells and macrophages/monocytes are the
main detectors of TLR7/8 agonists amongst other immune
blood cells (1). Because the route of siRNA delivery in vivo is
intrinsically related to a potential recruitment of immune blood
cells, it is important to assess the detection of siRNAs by TLR7/8
when selecting appropriate siRNA candidates for in vivo delivery.
Although uridine-based motifs within small RNA sequences have
been found to be important for TLR7/8 activation (4–7), in
silico prediction of the overall immunostimulatory potency of an
siRNA remains highly inaccurate. We and others have found
that single-stranded RNAs bearing uridine motifs that induce
strong immunostimulation in human PBMCs can be completely
masked when present in a double-stranded siRNA structure (6, 13).
For this reason, direct measurement of the immunogenicity of a
novel siRNA sequence is currently the most accurate method of
evaluating recruitment of TLR7/8 by siRNAs.
While both human TLR7 and TLR8 (hTLR7/8) have been
implicated in sequence-specific sensing of small RNAs, the
murine homolog of TLR8 is not able to detect RNA on its own
(12, 15, 16). Rather, sequence-specific sensing of RNAs relies
exclusively on TLR7 in the mouse (12, 15). It has recently been
shown by us and others that hTLR7 and hTLR8 recognize dif-
ferent RNA sequences, thus the immunogenicity of some
sequences preferentially recognized by hTLR8 is not conserved
between human and mouse (12, 15, 16). Nevertheless, our
observations based on a large panel of oligoribonucleotides have
led us to the conclusion that sequence-specific sensing of small
RNAs by TLR7 is conserved between human and mouse (12)
(see Fig. 1). Here, we describe two protocols allowing for the
evaluation of mouse TLR7 (and per se, also human TLR7) and
human TLR8 recruitment by siRNAs. For mouse TLR7 recruit-
ment, we rely on the induction of mouse TNF-a (mTNF-a) by
a macrophage-like cell line (RAW 264.7) (5, 12). Making use of
the conservation of TLR7 sensing between human and mouse
avoids using a costly human interferon-a (IFN-a) ELISA and
yet captures most of the hTLR7-driven IFN-a response observed
in human PBMCs (see Fig. 1). It is noteworthy that when a
sequence is found not to trigger TNF-a induction in RAW 264.7
cells, no conclusion can be drawn regarding its innate immune
activating potential in human blood without further validation
of hTLR8 activity via human TNF-a (hTNF-a) production in
human PBMCs (see Fig. 1).
3. Methods
3.1. Sequence-Specific
Recruitment of TLR7
and 8
24

Monitoring Innate Immune Recruitment by siRNAs in Mammalian Cells
Plate RAW 264.7 cells passaged on surface-treated plasticware to
a confluency of ~80,000 cells per well of a 96-well plate in 150 mL
complete RPMI medium in the morning of the TLR stimulation
(see Note 1). Incubate the cells at 37°C in 5% CO
2
for a mini-
mum of 4 h prior to treatment with the TLR agonists.
3.1.1. Preparation
of Mouse RAW Cells
for TLR7 Activation
si
R
N
A1
si
R
N
A2
si
R
N
A3
si
R
N
A4
si
R
N
A5
M
oc
k
CL
75
M
ed
iu
m
0
250
500
750
1000
1250
1500
2000
m
TN
Fα

(pg
/m
l)
IF
N
α
(pg
/m
l)
si
R
N
A1
si
R
N
A2
si
R
N
A3
si
R
N
A4
si
R
N
A5
M
oc
k
CL
75
M
ed
iu
m
250
200
150
100
50
0
3000
2000
1000
0
hTN
Fα
(pg/m
l)
IFNα
TNFα
a
b
Fig. 1. siRNA-induced TNF-a in human and mouse macrophages. (a) Mouse RAW 264.7
cells and (b) human PBMCs were treated as presented in Subheading 3.1 with 750 nM
of siRNAs complexed with DOTAP for 18 h. Each treatment was carried out in biological
triplicate and the data is from one representative experiment for both (a) and (b). The
error bars represent the standard error of the mean (SEM). In this example, the mouse
macrophage cell line data (a) indicates that siRNA1, 3 and 4 are immunostimulatory
(through mouse TLR7), whereas siRNA2 and 5 are not. While a similar observation can
be made in human PBMCs (b) when looking at IFN-a (indicative of human TLR7 recruit-
ment), we find that siRNA2 is a good inducer of TNF-a (indicative of human TLR8
recruitment) but not IFN-a. However, siRNA5 appears to be a very low inducer of both
IFN-a and TNF-a in PBMCs and would therefore be considered here as very poorly
immunostimulatory
25