Journal of Immune Based Therapies, Vaccines and Antimicrobials, 2012, 1, 1-6
http://dx.doi.org/10.4236/jibtva.2012.11001 Published Online April 2012 (http://www.SciRP.org/journal/jibtva)
Poly-I:C Decreases Dendritic Cell Viability Independent of
PKR Activation
Hjalte List Larsen, Anders Elm Pedersen
Department of International Health, Immunology and Microbiology, The Panum Institute,
University of Copenhagen, Copenhagen, Denmark
Received March 1, 2012; revised April 2, 2012; accepted April 12, 2012
ABSTRACT
Vaccination with tumor-antigen pulsed, monocyte-derived dendritic cells (DCs) has emerged as a promising strategy in
cancer immunotherapy. The standard DC maturation cocktail consists of a combination of tumor necrosis factor-α
(TNF-α)/interleukin (IL)-1β/IL-6 and prostaglandin E2 (PGE2) for generation of standard DCs (sDCs). In order to im-
prove IL-12p70 production and cytotoxic T-lymphocyte (CTL) induction, a novel cocktail composed of TNF-α/IL-1β/
interferon (IFN)-α/IFN-γ and polyinosinic:polycytidylic acid (Poly-I:C) has been introduced to generate so-called
α-Type-1 polarized DCs (αDC1s). We and others have previously performed a comprehensive comparison of sDCs and
αDC1s. Here we demonstrate that the viability of αDC1s is lowered compared to sDCs and that DC apoptosis is medi-
ated by Poly-I:C. We speculated that activation of protein kinase R (PKR) could mediate the observed apoptosis, but
despite significantly higher PKR expression in αDC1s compared to sDCs and induction of active threonine (Thr)446
autophosphorylation of PKR in αDC1s, Poly-I:C did not influence total PKR expression or autophosporylation, indi-
cating PKR-independent Poly-I:C-induced DC apoptosis.
Keywords: Cancer Immunotherapy; Dendritic Cells; Poly-I:C; PKR
1. Introduction
In therapeutic cancer vaccination with monocyte-derived
DCs [1], maturation is mandatory for proper T-cell
stimulation [2]. DCs maturated according to the standard
cytokine cocktail consisting of TNF-α/IL-1β/IL-6 and
PGE2 [3] show high expression of maturation markers
CD80, CD83, CD86, CD134L and CCR7, but low secre-
tion of IL-12p70 [4] which may have negative conse-
quences for activation of appropriate T cell responses
such as TH1 cells and cytotoxic T lymphocytes (CTLs).
In the search of cytokine cocktails that increase DC
IL-12p70 secretion, DCs maturated according to the
α-Type-1 polarizing cocktail containing TNF-α/IL-1β/
Poly-I:C/IFN-α and IFN-γ were developed, as these DCs
produce large amounts of IL-12p70 [4,5]. We and others
have characterized cytokine secretion and capacity for T
cell activation in details elsewhere [4,5] and it has been
found that αDC1s may have a reduced capacity for T cell
activation under certain conditions [4]. In addition to the
criterias for successful DC generation investigated else-
where, viability is an important factor. Here, we report a
decreased viability of αDC1s and investigate a potential
role of PKR mediated mechanisms.
The double-stranded RNA (dsRNA)-analog Poly-I:C
included in the α-Type-1 polarizing cocktail is a potential
inducer of apoptosis, as cytosolic Poly-I:C is capable of
activating the dsRNA-activated protein kinase (PKR) [6].
Upon dsRNA/Poly-I:C binding, PKR is autophosphory-
lated on multiple sites including Thr446 and Thr451 [7],
enabling signal transduction which can induce global
translation inhibition and apoptosis [8]. Furthermore,
Poly-I:C can be detected by toll-like receptor 3 (TLR3)
in the endosomal compartment of DCs [9] and it has
been reported that PKR is recruited to a large signaling
complex following Poly-I:C binding to TLR3 [10]. We
thus speculated that Poly-I:C is the cause of lowered
αDC1 viability as compared to sDCs, and that such a Poly-
I:C-mediated induction of apoptosis was mediated by
PKR activation. In this case, siRNA mediated silencing
of PKR-dependent apoptotic pathways could be a future
strategy for improved efficacy of therapeutic αDC1s.
Here, we report that absence of Poly-I:C from the
α-Type-1 polarizing cocktail significantly improved αDC1
viability to levels comparable to sDCs without signifi-
cantly decreasing CD80, CD83 and CD86 expression.
αDC1s expressed significantly higher amounts of PKR
than iDCs and sDCs, with PKR expression being de-
pendent on the presence of IFNs in the maturation cock-
tail but not on Poly-I:C. We were also unable to detect an
C
opyright © 2012 SciRes. JIBTVA

H. L. LARSEN ET AL.
2
effect of Poly-I:C on PKR autophosphorylation in mature
DCs, suggesting that the observed lowered viability of
αDC1s was PKR-independent.
2. Methods
2.1. Generation of Human Monocyte-Derived
DCs
Human monocyte-derived DCs were generated as previ-
ously described with GM-CSF + IL-4 [11] from buffy
coats after informed consent from healthy donors at the
local department of clinical immunology. Maturation of
iDCs was performed on day 7. sDCs were maturated
with TNF-α (10 ng/mL), IL-1β (10 ng/mL), IL-6 (10
ng/mL) (all cytokines from Peprotech) and 1 μg/mL
PGE2 (Prostin E2, Pfizer). The αDC1 phenotype was
induced by TNF-α (10 ng/mL), IL-1β (10 ng/mL), IFN-α
(10 ng/mL), IFN-γ (10 ng/mL) (all cytokines from
Peprotech) and 20 μg/mL Poly-I:C (Sigma Aldrich). DCs
were maturated 48 hours prior to analysis.
2.2. Flow Cytometry
Flow cytometry was performed using antibodies against
the following surface proteins: CD11c, CD80, CD83 and
CD86 (BD-Pharmingen). Isotope control antibodies were
obtained from DAKO. Detection of apoptotic DCs were
performed using annexin V-FITC apoptosis detection Kit
I (BD Pharmingen) according to the manufacturer’s pro-
tocol. Data acquisition and analysis was performed on a
FACSCalibur flow cytometer (BD) using CellQuest
software (BD Biosciences). DCs were identified as the
large granular cell population based on forward scatter
(FSC)/side scatter (SSC) linear dot plots.
2.3. Western Blotting
DCs were lysed in RIPA buffer supplemented with pro-
teinase inhibitor cocktail (Sigma Aldrich), PMSF, NaF
and Na3VO4 during 20 minutes of centrifugation at
14,000 g, 4˚C. Lysates were treated with SDS and heated
for 5 minutes prior to loading onto 4% - 12% Bis-Tris
gels (Invitrogen). Electrophoresis was performed in MES
running buffer (Invitrogen) for 1 hour at 200 V. Blotting
was performed at 30 V for 1 hour onto nitrocellulose
membranes (Invitrogen) in transfer buffer (Invitrogen).
Membranes were blocked and probed with polyclonal
goat anti-pPKR (Thr446) antibody, polyclonal rabbit
anti-PKR antibody (both Santa Cruz Biotechnology, Inc.)
or goat anti-GAPDH antibody (AbCam) over night at 4˚C.
Secondary horse radish peroxidase (HRP)-conjugated
polyclonal antibodies (DAKO) were added and the
membrane was incubated 1 hour at room temperature.
The membrane was washed and visualized using an en-
hanced chemiluminescence detection system (ECL plus
Western Blotting Detection system, GE Healthcare) ac-
cording to the manufacturer’s instructions. Detection of
HRP-generated precipitates was performed on a Typhoon
Scanner 9410 with Typhoon Scanner software (both from
Amersham Biosciences). The same software was used to
analyze band intensity, with PKR and pPKR intensity
being normalized to GAPDH expression.
3. Results
3.1. DC Phenotype after Maturation
Day 7 IL-4 and GM-CSF differentiated monocyte-de-
rived DCs were identified as large granular cells ex-
pressing the DC marker CD11c (Figures 1(A) and (B)).
Maturation using the two established cocktails for gen-
eration of sDCs and αDC1s both significantly increased
the expression of CD80 (p < 0.05 and p < 0.01 respec-
tively) and CD83 (p < 0.05 and p < 0.05 respectively)
compared to iDCs, while a stable high expression level
of CD86 was maintained from iDCs into mature DCs.
CD83 expression in sDC was less pronounced in these
DCs from healthy donors as compared to our previous
results from cancer patient derived DCs [4]. Removal of
individual cytokine components from the α-Type-1 po-
larizing cocktail did not induce any significant changes
in the expression of the three maturation markers, how-
ever absence of Poly-I:C or IFN-α reduced the expres-
sion of both CD80 and CD83 compared to expression
levels in αDC1s. These observed reductions in CD80 and
CD83 expression were not statistical significant but con-
sistent in all donors. It was previously published that
IL-12p70 secretion was dependent on IFN-γ or Poly-I:C
with the largest contribution from IFN-γ [5]. For T cell
activation, αDC1s have been shown to be superior for
CTL activation [5], whereas assays mostly reflecting
CD4+ T cell responses has demonstrated a decreased
capacity for T cell activation as compared to sDCs [4].
3.2. Viability of ex Vivo Generated DCs
Annexin V-FITC and propidium iodide (PI) staining re-
vealed a tendency toward that sDCs were more viable
(mean of all donors: 79.4% living cells) as compared to
αDC1s (mean of all donors: 71.7% living cells), a dif-
ference that was not statistical significant (p = 0.069) but
consistent in all donors. Data from one donor is shown in
Figure 2(A) and mean of all donors in Figure 2(B). In
order to identify the cause of the viability changes in
αDC1s, all single components from the cocktail was sys-
tematically excluded (Figure 2(B)). Exclusion of Poly-I:C
from the α-type-1 polarizing cocktail significantly im-
proved the viability of DCs from a mean of 71.7%
(αDC1s) to 79.9% living cells (p = 0.02), thus increasing
DC viability to a level comparable to sDCs. The viability

H. L. LARSEN ET AL.
3
(A) (B)
(C)
Figure 1. DC Phenotype. (A) DCs were initially identified as the large granular population based on flow cytometry FSC/SSC
plots. Flow cytometric analysis was performed in order to determine the surface expression of the maturation markers CD80,
CD83 and CD86 on CD11c + DCs; (B) The percentage of CD11c+ DCs staining positive of individual maturation markers in
one representative of 4 donors. a: iDCs, b: sDCs: c: αDC1s, d: αDC1s – Poly-I:C, e: αDC1s – TNF-α, f: αDC1s – IL-1β, g:
αDC1s – IFN-α, h: αDC1s – IFN-γ; (C) Mean percentage of maturation marker positive CD11c+ DCs maturated according to
the indicated cytokine cocktails (n = 4). (*) indicates significant different maturation marker expression compared to iDCs (p
< 0.05), two-sided Students t-test.
of DCs maturated with all other α-Type-1 polarizing
cocktail derivatives containing Poly-I:C was also reduced
compared to sDCs and αDC1s without Poly-I:C (Figure
2(B)). Exclusion of other cytokines from the α-Type-1
polarizing cocktail did not significantly alter the per-
centage of viable DCs. However, inclusion of Poly-I:C in
a maturation cocktail does not seem to reduce DC capa-
city for overall in-vitro allogeneic T cell stimulation
(data not shown).
3.3. Poly-I:C Mediated Effects on PKR
Expression and Activation
Reports of PKR activation upon direct binding of cytoso-
lic Poly-I:C, and recruitment of PKR to a signaling com-
plex binding to the cytosolic tail of TLR3 following
Poly-I:C binding in the endosomal compartment, led to
the speculation that the apparent Poly-I:C mediated re-
duction in DC viability might be caused by activation of
PKR. Here we demonstrate that PKR expression in sDCs
was not elevated as compared to iDCs (Figures 3(A) and
(B)). In contrast, PKR expression was elevated in αDC1s
as compared to iDCs and sDCs (p < 0.05). However, no
significant difference in PKR expression was found be-
tween DCs maturated using the αDC1 cocktail with (14.7
fold increase) or without Poly-I:C (15.2 fold increase),
suggesting that Poly-I:C has no effect on total DC PKR
expression. In contrast, PKR expression of αDC1s was
dependent on IFN-α- (p < 0.05) (Figure 3(A) and (B)).
Other subtractions did not lead to significant changes as
compared to PKR expression in αDC1s.
As the role of Poly-I:C in PKR dynamics was expected
to be found at the level of activation rather than on tran-
scriptional regulation we next tested the levels of PKR
Thr446 autophosphorylation in DCs and how absence of
Poly-I:C affected PKR activation. Here, αDC1 matura-
tion resulted in autophosphorylation in 2 out of 4 donors,
but we were unable to detect any differences in the levels

H. L. LARSEN ET AL.
4
(A)
(B)
Figure 2. Exclusion of Polyα-Type-1 polariz-
f Thr446 phosphorylated PKR in αDC1s and αDC1s
4. Discussion
uality of ex vivo generated DCs for
-I:C from the
ing cocktail increases αDC1 viability. DCs maturated with
the indicated cytokine cocktails were stained with annexin
V FITC and PI, and subjected to flow cytometric analysis.
(A) Dot plots of annexin V FITC and PI stained DCs from
one representative of 8 donors. The percentage of viable
cells is indicated in the lower left corner; (B) Mean per-
centage of viable cells from all donors (n = 8). (*) indicates
significant different levels of viability (p < 0.05), one-sided
Students t-test.
o
maturated without Poly-I:C for 24 or 48 hours (Figure 3(C)
and data not shown) or during the first 6 hours after Poly-
I:C addition to DCs maturated with IFN-α (data not shown).
Improving the q
(A)
(B)
(C)
Figure 3. PKR expression is unaffected by
erapeutic cancer vaccination is of paramount impor-
and activation
extracellular Poly-I:C. PKR expression was investigated by
western blotting in DCs maturated with the indicated cyto-
kine cocktails for 48 hours. (A) One representative blot is
shown (n = 3); (B) PKR expression levels were normalized
to GAPDH levels and the mean fold change in PKR expres-
sion compared to iDCs is plotted (n = 3). (*) indicates sig-
nificant different PKR expression (p < 0.05), two-sided
Students t-test; (C) Active Thr446 phosphorylated PKR
detection by western blotting was performed 24 and 48
hours after addition of the indicated maturation cocktails to
iDCs. 2 out of 4 donors demonstrated Thr446 phosphory-
lated PKR. A western blot from one of these two donors is
shown.
th
tance and may increase the efficiency of this vaccine
strategy. αDC1s have been shown to be superior in
IL-12p70 secretion and CTL activation [5]. Viability of
therapeutic DCs is another factor of potential importance
as DCs may exert their effect in vivo after more than 24
hours upon injection. In order to improve αDC1 yields,
we tried to identify the cause of Poly-I:C mediated DC
death, in the attempt of identifying a target for future
siRNA-mediated silencing strategies as a mean of gener-
ating functional αDC1s with increased viability. In par-
ticular we investigated the potential role of PKR induc-
tion and autophosphorylation.

H. L. LARSEN ET AL. 5
Poly-I:C-mediated IL-12p70 secretion and induction
of
n of PKR in αDC1s
m
pression levels in αDC1s, we ob-
se
sion, we identified a Poly-I:C dependent
lo
5. Acknowledgements
the Faculty of Health Sci-
REFERENCES
[1] F. Sallusto anicient Presentation
type I IFNs in αDC1s [5] makes this TLR3 ligand in-
dispensable in the α-Type-1 polarizing cocktail, and thus
removal of Poly-I:C from the cytokine cocktail in order
to restore DC viability is undesirable. However, apop-
tosis of DCs generated for therapeutic cancer vaccination
is undesirable due to reduced DC yields from valuable
and limited DC precursor cells and also decreases lon-
gevity upon in vivo injection. Furthermore, the presence
of apoptotic cells can impair DC phenotypic maturation
by suppressing the expression of co-stimulatory mole-
cules [12], hereby impairing DC functionality and ulti-
mately vaccine efficacy. However, this does not seem to
occur at the percentage of dead cells occurring in our
system (Figures 1(B) and (C)).
We hypothesized that activatio
ight cause the apparent Poly-I:C mediated cell death, as
active PKR induces translational inhibition and apoptosis
as a defense mechanism against viral infection [7,8].
Characterization of PKR expression in DCs revealed
significantly higher amounts of PKR in αDC1s and
αDC1s maturated without Poly-I:C as compared to sDCs
(Figure 3). The finding that presence of Poly-I:C has no
effect on PKR expression is in compliance with the
known role of Poly-I:C as a PKR activator and not as an
inducer of PKR transcription [6]. The high PKR expres-
sion observed in αDC1s seems to be induced by IFNs, as
αDC1s maturated without IFN-α showed a significantly
reduced PKR expression as compared to αDC1s. Matura-
tion of αDC1s without IFN-γ also seemed to reduce PKR
expression as compared to αDC1s however this reduced
expression was non-significant. These results confirm the
established role of IFNs as inducers of PKR expression
(reviewed in [13]).
Despite high PKR ex
rved no significant difference in the amounts of active
Thr446 phosphorylated PKR in αDC1s and αDC1s
maturated without Poly-I:C (Figure 3(C)). The observed
lack of Poly-I:C mediated PKR activation at neither 48
hours after addition nor during the first 6 hours (data not
shown) suggests that PKR is not activated by extracellu-
lar Poly-I:C. These results argue against a role of acti-
vated PKR as the inducer of cell death in αDC1s despite
our previous findings of Poly-I:C as the cause of the
lowered viability of αDC1s. The apparent discrepancy
regarding the ability of Poly-I:C to induce Thr446 PKR
autophosphorylation in our hands and others might be
due to Poly-I:C residing in different compartments. PKR
activation by Poly-I:C reported by McAllister et al. was
observed after transfection, hereby enabling direct cyto-
solic binding of Poly-I:C to PKR [6], whereas Poly-I:C
was added extracellularly in our protocol for DC matura-
tion. We were unable to find any reports of translocation
of Poly-I:C from the endosomal compartment to the cy-
toplasm of DCs, thus explaining the lack of PKR activa-
tion. Furthermore, PKR activation following recruitment
to a TLR3 signaling complex in the endosome, activated
upon Poly-I:C addition was not directly detected by Jiang
et al., as the effect of transfection with dominant negative
PKR was only investigated on the level of NF-κB activ-
tion [10].
In conclu
wered viability of αDC1s where the mechanisms of
apoptosis was independent of PKR expression and acti-
vation. Future studies may reveal the targets for this
Poly-I:C mediated apoptosis which could be exploited
for siRNA silencing, thus maintaining the efficient αDC1
mediated maturation without hampering DC viability as
compared to sDCs.
This work was supported by
ences, University of Copenhagen and a grant from The
Augustinus Foundation.
d A. Lanzavecchia, “Eff
of Soluble Antigen by Cultured Human Dendritic Cells Is
Maintained by Granulocyte/Macrophage Colony-Stimula-
ting Factor plus Interleukin 4 and Downregulated by
Tumor Necrosis Factor Alpha,” The Journal of Experi-
mental Medicine, Vol. 179, No. 4, 1994, pp. 1109-1118.
doi:10.1084/jem.179.4.1109
[2] I. J. de Vries, W. J. Lesterhuis, N. M. Scharenborg, L. P.
Steinbrink, L.
Engelen, D. J. Ruiter, M. J. Gerritsen, et al., “Maturation
of Dendritic Cells Is a Prerequisite for Inducing Immune
Responses in Advanced Melanoma Patients,” Clinical
Cance Research, Vol. 9, 2003, p. 5091.
[3] H. Jonuleit, U. Kuhn, G. Muller, K.
Paragnik, E. Schmitt, et al., “Pro-Inflammatory Cytokines
and Prostaglandins Induce Maturation of Potent Immuno-
stimulatory Dendritic Cells under Fetal Calf Serum-Free
Conditions,” European Journal of Immunology, Vol. 27,
No. 12, 1997, pp. 3135-3142.
doi:10.1002/eji.1830271209
[4] R. Trepiakas, A. E. Pedersen, O. Met, M. H. Hansen, A.
Berntsen and I. M. Svane, “Comparison of Alpha-Type-1
Polarizing and Standard Dendritic Cell Cytokine Cocktail
for Maturation of Therapeutic Monocyte-Derived Den-
dritic Cell Preparations from Cancer Patients,” Vaccine,
Vol. 26, No. 23, 2008, pp. 2824-2832.
doi:10.1016/j.vaccine.2008.03.054
[5] R. B. Mailliard, A. Wankowicz-Kalinska, Q. Cai, A.
Wesa, C. M. Hilkens, M. L. Kapsenberg, et al., “Alpha-
Type-1 Polarized Dendritic Cells: A Novel Immunization
Tool with Optimized CTL-Inducing Activity,” Cancer
Research, Vol. 64, No. 17, 2004, pp. 5934-5937.
doi:10.1158/0008-5472.CAN-04-1261
[6] C. S. McAllister and C. E. Samuel, “The RNA-Activated
Protein Kinase Enhances the Induction of Interferon-Beta

H. L. LARSEN ET AL.
6
and Apoptosis Mediated by Cytoplasmic RNA Sensors,”
Journal of Biological Chemistry, Vol. 284, No. 3, 2009,
pp. 1644-1651. doi:10.1074/jbc.M807888200
[7] F. Zhang, P. R. Romano, T. Nagamura-Inoue, B. Tian, T.
E. Dever, M. B. Mathews, et al., “Binding of Double-
Stranded RNA to Protein Kinase PKR Is Required for
Dimerization and Promotes Critical Autophosphorylation
Events in the Activation Loop,” Journal of Biological
Chemistry, Vol. 276, No. 27, 2001, pp. 24946-24958.
doi:10.1074/jbc.M102108200
[8] D. Scheuner, R. Patel, F. Wang, K. Lee, K. Kumar, J. W
et al., “Double-Stranded RNA-
u
Dependent Protein Kinase
,
Phosphorylation of the Alpha-Subunit of Eukaryotic
Translation Initiation Factor 2 Mediates Apoptosis,”
Journal of Biological Chemistry, Vol. 281, No. 30, 2006,
pp. 21458-21468. doi:10.1074/jbc.M603784200
[9] L. Alexopoulou, A. C. Holt, R. Medzhitov and R. A.
Flavell, “Recognition of Double-Stranded RNA and
Activation of NF-kappa B by Toll-Like Receptor 3,”
Nature, Vol. 413, No. 6857, 2001, pp. 732-738.
doi:10.1038/35099560
[10] Z. Jiang, M. Zamanian-Daryoush, H. Nie, A. M. S
R. Williams and X. Li,
ilva, B.
“Poly(I-C)-Induced Toll-Like
Receptor 3 (TLR3)-Mediated Activation of NFkappa B
and MAP Kinase Is through an Interleukin-1 Receptor-
Associated Kinase (IRAK)-Independent Pathway Employ-
ing the Signaling Components TLR3-TRAF6-TAK1-
TAB2-PKR,” Journal of Biological Chemistry, Vol. 278,
No. 19, 2003, pp. 16713-16719.
doi:10.1074/jbc.M300562200
[11] A. E. Pedersen, M. Thorn, M. Gad, M. R. Walter, H. E.
Johnsen, E. Gaarsdal, et al., “Phenotypic and Functional
Characterization of Clinical Grade Dendritic Cells
Generated from Patients with Advanced Breast Cancer
for Therapeutic Vaccination,” Scandinavian Journal of
Immunology, Vol. 61, No. 2, 2005, pp. 147-156.
doi:10.1111/j.0300-9475.2005.01531.x
[12] C. A. Williams, R. A. Harry and J. D. McLeod,
.x
“Apoptotic Cells Induce Dendritic Cell-Mediated Suppres-
sion via Interferon-Gamma-Induced IDO,” Immunology,
Vol. 124, No. 1, 2008, pp. 89-101.
doi:10.1111/j.1365-2567.2007.02743
an, “The dsRNA [13] M. A. Garcia, E. F. Meurs and M. Esteb
Protein Kinase PKR: Virus and Cell Control,” Biochimie,
Vol. 89, No. 6-7, 2007, pp. 799-811.
doi:10.1016/j.biochi.2007.03.001