Original article 351
Temporally designed treatment of melanoma cells by ATRA and polyI:C results in enhanced chemokine and IFNb secretion controlled differently by TLR3 and MDA5 Attila Szaboa,b, Rolah M. Osmana, Ildiko Bacskaia,b, Brahma V. Kumara,b, Zsofia Agoda,b, Arpad Lanyia,b, Peter Gogolaka,b and Eva Rajnavolgyia,b In the last three decades, the incidence of melanoma has increased worldwide and no effective treatment modalities have been developed yet. All-trans retinoic acid (ATRA) and polyinosinic:polycytidylic acid (polyI:C) are strong inducers of toll-like receptor 3 (TLR3) and MDA5 expression, and polyI:C-induced TLR3 and MDA5 signaling specifically causes cell death in melanoma cells in vitro. We addressed the question of whether ATRA pretreatment could enhance the efficacy of polyI:C and, if so, would ATRA have any additional effects on this process. We found that the combined treatment of human melanoma cells with ATRA and polyI:C strongly increased the expression of TLR3 and MDA5 in both WM35 and WM983A cells associated with significantly higher mRNA and secreted levels of interferon b (IFNb), CXCL1, CXCL8/IL-8, CXCL9, and CXCL10 than cells treated with either ATRA or polyI:C. Silencing of MDA5 by siRNA moderately affected IFNb secretion, whereas TLR3 knockdown interfered with both CXCL chemokine and IFNb production. Furthermore, the supernatants of ATRA + polyI:C-activated cultures increased the migration of both human monocyte-derived macrophages and CD1a + dendritic cells significantly as
Introduction Melanoma is a malignant tumor of melanocytes that causes the majority (75%) of skin cancer-related deaths. Its incidence has increased significantly in the last three decades, and no effective treatments have been found yet [1]. Besides standard chemotherapy, cytokines have also been used for melanoma treatment [2]. Adjuvant cytokine therapies [3,4], such as concomitant type I interferon (IFN) administration, are novel and promising approaches that mobilize professional antigen-presenting cells (APCs) and cytotoxic T lymphocytes (CTLs) [1,5]. The skin is the body’s first line of defense against invading microbes and physical and chemical insults. Constitutive immunosurveillance of this large and exposed organ requires properly controlled response mechanisms carried out by immune sentinels and effector cells [6]. Skin resident professional APCs, such as dermal macrophages (MACs) and dendritic cells (DCs), play important roles in the regulation of defense mechanisms. However, the key players in this process are the epidermal Langerhans cells, a subtype of DCs distinguished by a high expression of c 2012 Wolters Kluwer Health | Lippincott Williams & Wilkins 0960-8931
compared with the supernatants of cells treated with either ATRA or polyI:C, and this effect occurred in a TLR3-dependent manner. In conclusion, consecutive treatment with ATRA and polyI:C results in strong, TLR3/MDA5-mediated chemokine and IFN responses in cultured human melanoma cells, which triggers a functional migratory response in professional antigen-presenting cells. This novel mode of concomitant activation may represent a more efficient treatment option for future melanoma therapy. Melanoma Res c 2012 Wolters Kluwer Health | Lippincott 22:351–361 Williams & Wilkins. Melanoma Research 2012, 22:351–361 Keywords: chemokines, MDA5, toll-like receptor 3, type I interferon a
Department of Immunology and bResearch Centre for Molecular Medicine, University of Debrecen Medical and Health Science Center, Debrecen, Hungary Correspondence to Attila Szabo, Department of Immunology, University of Debrecen Medical and Health Science Center, Egyetem sqr. 1, 4032 Debrecen, Hungary Tel/fax: + 36 524 17159; e-mail:
[email protected] Received 23 April 2012 Accepted 15 June 2012
CD207/langerin and CD1a [7]. Many skin-related effector cells including CD4 + T-helper cells, innate gd T and natural killer T cells, mast cells, and fibroblasts are also involved [6]. APCs recognize both microbial and tumorderived antigens and subsequently present them to T cells in the draining lymph node, thereby linking innate and adaptive immunity. Epidermal and dermal APCs express several pattern recognition receptors (PRRs) that detect both endogenous and exogenous ligands. The recognition of specific motifs by PRRs activates complex signaling cascades and may elicit proinflammatory and IFN responses. These responses depend on the delicate balance of the activation kinetics and cross-talk of different PRRs [8,9]. Toll-like receptors (TLRs) and RIG-I-like receptors are important PRR family members that mediate nuclear factor (NF)-kB-dependent inflammatory cytokine-dependent, chemokine-dependent, and interferon regulatory factor (IRF)-dependent IFN responses, respectively [10]. The production of type I IFNs depends on the phosphorylation and the subsequent nuclear translocation of IRFs. This branch of PRR-mediated signaling represents an early and rapid innate response that induces IFNa and IFNb DOI: 10.1097/CMR.0b013e328357076c
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
352 Melanoma Research 2012, Vol 22 No 5
production. IFNb is one of the major effector cytokines that couples innate and adaptive immunity and mediates strong antiviral and antitumor responses by activating many different cell types [11,12]. The NF-kB pathway controls the gene expression of cytokines, crucial in the regulation of inflammation and adaptive immune responses, such as interleukin-1b (IL-1b), IL-6, tumor necrosis factor a (TNF-a), IL-10, and IL-12. NF-kB also controls the production of several chemokines secreted by activated APCs and keratinocytes. These include CXCL1, CXCL8/ IL-8, CXCL9, and CXCL10/IP-10, which modulate the migration of different cell types into the skin [6]. In detail, CXCL1 and CXCL8 recruit neutrophils, whereas CXCL9 and CXCL10 attract activated T lymphocytes, MACs, and DCs to inflamed tissues and the epidermis in a CXCR3dependent manner [13–16]. CXCLs have also been shown to be either beneficial or deleterious in melanoma tumor progression and metastasis [17–19]. In particular, the IFNinduced CXCL10 can effectively reduce the proliferation and invasivity of melanoma cell both in vitro and in vivo [19]. However, the interplay of this complex network of cytokines and chemokines in melanoma biology has yet to be understood. The synthetic dsRNA analog polyinosinic:polycytidylic acid (polyI:C) was described as a specific ligand of TLR3 and MDA5 [20,21]. Triggering of these PRRs by polyI:C activates both the NF-kB and the type I IFN pathways and leads to the production of proinflammatory cytokines, chemokines, and IFNs. Recently, polyI:C has emerged as a potent adjuvant in cancer immunotherapy [22,23]. As melanoma cells express both TLR3 and MDA5, polyI:C has become a possible treatment for this cancer type [24,25]. Indeed, ligation of TLR3 and MDA5 by polyI:C was shown to convey proapoptotic and antiproliferative signals to melanoma cells [26,27]. This effect is initiated by the mitochondrial apoptotic pathway and includes Apaf-1 and caspase-9 [24]. Moreover, peptide vaccination with the concomitant administration of polyI:C could also augment the expansion of antigenspecific CTLs and the generation of memory CD8 + T cells, leading to efficient elimination of malignant cells [28]. All-trans retinoic acid (ATRA) is another therapeutic agent that has been used as an adjuvant in melanoma chemotherapy [29]. ATRA inhibits cell growth and differentiation and was shown to disrupt mitochondrial functions, causing apoptosis in in-vitro cultured melanoma cells [30,31]. It was also reported that ATRA can downregulate the expression of vitronectin receptor, thereby preventing tumor cell invasion and dissemination [32]. Thus, polyI:C and ATRA are emerging as promising drug candidates in melanoma therapy. In this study, we examined whether a temporally designed treatment of melanoma cells with ATRA and polyI:C could enhance the immunomobilizing capacity of these compounds. To assess the efficacy of ATRA pretreatment, we tested the inflammatory cytokine,
chemokine, and type I IFN secretion of two human melanoma cell lines triggered by polyI:C. We found that even a single 24h ATRA pretreatment significantly enhanced the activating capacity of polyI:C. We also aimed to determine the mechanisms behind these cytokine responses.
Methods Cell types, isolation, and phenotyping
WM35 and WM983A human melanoma cells were obtained from ATCC (Teddington, Middlesex, UK) and cultured according to the recommended protocol. Leukocyte-enriched buffy coats were obtained from healthy blood donors drawn at the Regional Blood Center of Hungarian National Blood Transfusion Service (Debrecen, Hungary) in accordance with the written approval of the Director of the National Blood Transfusion Service and the Regional and Institutional Ethics Committee of the University of Debrecen, Medical and Health Science Center (Debrecen, Hungary). Written informed consent was obtained from donors before blood donation and their data were processed and stored according to the directives of the European Union. PBMCs were separated by a standard density gradient centrifugation with Ficoll-Paque Plus (Amersham Biosciences, Uppsala, Sweden). Monocytes were purified from PBMCs by positive selection using immunomagnetic cell separation with anti-CD14 microbeads according to the manufacturer’s instructions (Miltenyi Biotec, Bergisch Gladbach, Germany). After separation on a VarioMACS magnet, 96–99% of the cells were CD14 + monocytes as measured by flow cytometry. For monocytederived dendritic cell (moDC) generation, monocytes were cultured in 12-well tissue culture plates at a density of 2 106 cells/ml in AIM-V medium (Invitrogen, Carlsbad, California, USA) supplemented with 80 ng/ml GM-CSF (Gentaur Molecular Products, Brussels, Belgium) and 100 ng/ml IL-4 (Peprotech EC, London, UK). On day 2, the same amounts of GM-CSF and IL-4 were added to the cell cultures, and moDCs were harvested on day 5. For monocyte-derived macrophage (moMAC) differentiation, 50 ng/ml M-CSF (Gentaur Molecular Products) was added to the monocyte cultures on days 0 and 2, and fully differentiated MACs were collected on day 4. Phenotyping of moDCs and moMACs was performed by flow cytometry using anti-CD209-PE, anti-CD1a-PE, anti-CD14-PE (Beckman Coulter, Hialeah, Florida, USA), anti-CD68-PE, anti-HLA-DR-FITC, and isotypematched control antibodies (Ab) (BD Biosciences, Franklin Lakes, New Jersey, USA). Fluorescence intensities were measured by FACS Calibur (BD Biosciences) and data were analyzed using the FlowJo software (Tree Star, Ashland, Oregon, USA). CD1a + moDCs were separated using the FACS DiVa high-speed cell sorter (BD Biosciences).
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
PRR-induced responses in melanoma cells Szabo et al. 353
Activation of melanoma cells
ATRA (Sigma, Schnelldorf, Germany) and high-molecular-weight polyI:C (InvivoGen, San Diego, California, USA) were used at working concentrations of 10 – 6 mol/l (ATRA) and 20 mg/ml (polyI:C). ATRA treatments were performed for 1, 2, or 3 days, whereas polyI:C activation of TLR3/MDA5 was performed for 24 h in all cases. To prepare cell lysates for western blotting, cells were activated for 24–48 h. To collect supernatants for ELISA and prepare cell lysates for quantitative PCR (Q-PCR), sample collection was performed 24 h after activation. RNA isolation, cDNA synthesis, and Q-PCR
A real-time Q-PCR was performed as described previously [33]. Briefly, total RNA was isolated by TRIzol reagent (Invitrogen). Total RNA (1.5–2 mg) were reverse transcribed using SuperScript II RNase H reverse transcriptase (Invitrogen) and Oligo(dT)15 primers (Promega, Madison, Wisconsin, USA). Gene-specific TaqMan assays (Applied Biosystems, Foster City, California, USA) were used to perform Q-PCR in a final volume of 25 ml in triplicate using AmpliTaq DNA polymerase and the ABI Prism 7900HT real-time PCR instrument (Applied Biosystems). Amplification of 36B4 was used as a normalizing control. Cycle threshold values were determined using SDS 2.1 software (Schawk Digital Solutions, Chicago, Illinois, USA). Constant threshold values were set for each gene throughout the study. The sequences of the primers and probes are available upon request. siRNA experiments
Gene-specific siRNA knockdown of TLR3 and MDA5 was performed by SilencerSelect siRNA (Applied Biosystems) transfection using the GenePulser Xcell instrument (BioRad, Hercules, California, USA). Silencer Negative Control nontargeting siRNA (Applied Biosystems) was used as a negative control. The efficacy of siRNA treatments was tested 3 days after transfection by Q-PCR and western blotting. Western blotting
Cells were lysed in Laemmli buffer and the protein extracts were tested by Ab specific for TLR3 (Abcam, Cambridge, UK), MDA5 (Lifespan, Seattle, Washington, USA), and b-actin (Sigma) all diluted 1 : 1000; secondary Ab were used at a dilution of 1 : 5000. Anti-rabbit Ab conjugated to horseradish peroxidase (GE Healthcare, Little Chalfont Buckinghamshire, UK) was used as the secondary Ab. The SuperSignal enhanced chemiluminescence system was used for probing target proteins (Thermo Scientific, Rockford, Illinois, USA). After the membranes had been probed for TLR3 or MDA5, they were stripped and reprobed for b-actin. Cytokine measurements
Culture supernatants (1 106 cells/1 ml medium) were harvested 24 h after activation and the concentrations of
IL-1b, IL-6, TNF-a, CXCL1, CXCL8/IL-8, CXCL9, and CXCL10 were measured using OptEIA kits (BD Biosciences). The level of secreted IFNb was measured using a Human Interferon beta ELISA Kit (Cell Sciences, Canton, Massachusetts, USA). Migration assay
moMACs and CD1a+ moDCs were suspended in migration medium (0.5% BSA in RPMI 1640) at 106 cells/ml. Transwell migration inserts (diameter 6.5 mm; pore size 5 mm) were obtained from Corning (Lowell, Massachusetts, USA). Supernatants of melanoma cultures (2 106 cells/ml) were added to the lower chambers in a final volume of 600 ml. When spontaneous transendothelial migration assays were performed, the migration medium in the lower chamber was RPMI 1640 (0.5% BSA). moMACs and CD1a + moDCs were added to the upper chamber in a final volume of 250 ml, and the chemotaxis assays were carried out for 24 h in 5% CO2 at 371C. At the end of the assay, the inserts were discarded and cells that migrated to the lower chamber were collected. Migrated cell numbers were counted by flow cytometry using polystyrene standard beads (Sigma). Statistics
One-way analysis of variance, followed by the Bonferroni post-hoc test was used for multiple comparisons. All analyses were performed using SPSS statistics software, version 17.0 (SPSS Inc., Chicago, Illinois, USA). Differences were considered to be statistically significant at P less than 0.05. Significance is indicated as P less than 0.05.
Results Enhanced stimulation of TLR3 and MDA5 expression in WM35 and WM983A cells with consecutive ATRA and polyI:C treatments
The expression levels of TLR3 and MDA5 in different types of melanoma tumors have already been characterized [24,25]. However, not all melanoma cells express functional PRR proteins [25]. Thus, we first examined the expression of TLR3 and MDA5 in WM35 and WM983A cells at baseline and in cells activated with ATRA and polyI:C. We found that TLR3 and MDA5 are expressed in both WM35 and WM983A cells at both mRNA and protein levels (Fig. 1). To trigger TRL3 and MDA5 in these cells specifically, we used 20 mg/ml polyI:C alone or in combination with 1 mmol/l ATRA. ATRA treatments were performed for 1, 2, or 3 days, and when pretreatments were applied, they were followed by a single 24h polyI:C activation. Nontreated melanoma cells and cells treated by polyI:C for 24 h were used as controls. The relative expression levels of TLR3 and MDA5 were comparable in the two cell lines at both baseline and after activation, and WM983A cells showed higher expressions of MDA5 (Fig. 1a and b). Interestingly, consecutive treatments with ATRA and polyI:C led to higher
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
354
Melanoma Research 2012, Vol 22 No 5
Fig. 1
(a)
(b)
MDA5
TLR3
WM983A
WM35 ATRA d3 + polyI:C
ATRA d3 + polyI:C
ATRA d2 + polyI:C
ATRA d2 + polyI:C
ATRA d1 + polyI:C
ATRA d1 + polyI:C
ATRA d3
ATRA d3
ATRA d2
ATRA d2
ATRA d1
ATRA d1
PolyI:C
PolyI:C
Ctrl
Ctrl 0
0.5
1 1.5 Relative expression (c) Ctrl
PolyI:C
2
ATRA d1
ATRA d2
0
0.05 0.1 Relative expression
0.15
ATRA ATRA ATRA ATRA d3 d1+polyI:C d2+polyI:C d3+polyI:C TLR3
WM35
MDA5
β-Actin
Ctrl
PolyI:C
ATRA d1
ATRA d2
ATRA ATRA ATRA ATRA d3 d1+polyI:C d2+polyI:C d3+polyI:C TLR3
WM983A
MDA5 β-Actin
Induction of TLR3 and MDA5 expression in WM35 and WM983A cells. Gene and protein expression profile of the receptors TLR3 and MDA5 in melanoma cell lines upon consecutive treatments with ATRA and polyI:C. (a, b) mRNA expression of MDA5 (a) and TLR3 (b) in WM35 (light gray bars) and WM983A (dark gray bars) cells. One-day, 2-day, or 3-day treatments with 10 – 6 mol/l ATRA (ATRA d1, d2, d3) were performed alone, or were followed by a single 24h treatment with 20 mg/ml polyI:C (see also the Methods section). Mean±SEM values of triplicate measurements of three independent experiments are presented. (c) TLR3 and MDA5 protein expressions in WM35 and WM983A cells treated as described above (a, b). The results of a representative western blot experiment out of four are shown. ATRA, all-trans retinoic acid; Ctrl, non-treated control; TLR3, toll-like receptor 3.
expression levels of TLR3 and MDA5 at both mRNA (Fig. 1a and b) and protein levels (Fig. 1c) as compared with the control ATRA or polyI:C treatments. A single ATRA pretreatment (24 h) with the subsequent addition of polyI:C resulted in high expressions of TLR3 and MDA5 comparable with 48- or 72-h pretreatments. Thus, short-term consecutive treatments with ATRA and polyI:C can induce strong upregulation of TLR3 and MDA5 expressions in both WM35 and WM983A melanoma cells. Cytokine and chemokine expression of melanoma cells can be induced efficiently by consecutive treatments with ATRA and polyI:C
Our results showed that consecutive treatments with ATRA and polyI:C increased the expression of the innate
sensors TLR3 and MDA5 at both mRNA and protein levels (Fig. 1). We hypothesized that ATRA would not only increase the levels of TLR3 and MDA5 but also enhance receptor-mediated functions such as signaling upon subsequent polyI:C stimulation. In our next experiments, we used polyI:C as a specific activator [20,21] of ATRA-pretreated cells and tested the relative mRNA and secreted levels of various cytokines and chemokines involved in the activation and chemoattraction of immune cells. We found that consecutive treatments by ATRA and polyI:C increased the gene expression of IL-1b, IL-6, IFNb, CXCL10, CXCL9, CXCL8, and CXCL1 to a greater extent than treatments with either ATRA or polyI:C separately (Fig. 2a). In agreement with the findings presented in Fig. 2a, the levels of secreted proinflammatory cytokines and CXCL
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
PRR-induced responses in melanoma cells Szabo et al. 355
Fig. 2
WM35
(a)
WM983A
IL-1β
IL-6
ATRA d3 + polyI:C ATRA d2 + polyI:C ATRA d1 + polyI:C ATRA d3 ATRA d2
ATRA d3 + polyI:C ATRA d2 + polyI:C ATRA d1 + polyI:C ATRA d3 ATRA d2 ATRA d1 PolyI:C Ctrl
ATRA d1 PolyI:C Ctrl 0
0.1 0.2 0.3 Relative expression
0.4
0
IFNβ
0.02
0.04 0.06 0.08 Relative expression
0.1
CXCL10 ATRA d3 + polyI:C ATRA d2 + polyI:C ATRA d1 + polyI:C ATRA d3 ATRA d2 ATRA d1 PolyI:C Ctrl
ATRA d3 + polyI:C ATRA d2 + polyI:C ATRA d1 + polyI:C ATRA d3 ATRA d2 ATRA d1 PolyI:C Ctrl 0
0.2 0.4 0.6 Relative expression
0.8
0
0.1 0.2 Relative expression
0.3
CXCL8/IL-8
CXCL9 ATRA d3 + polyI:C ATRA d2 + polyI:C ATRA d1 + polyI:C ATRA d3 ATRA d2 ATRA d1 PolyI:C Ctrl
ATRA d3 + polyI:C ATRA d2 + polyI:C ATRA d1 + polyI:C ATRA d3 ATRA d2 ATRA d1 PolyI:C Ctrl 0
0.005 0.01 0.015 Relative expression
0.02
0
0.5 1 Relative expression
1.5
CXCL1 ATRA d3 + polyI:C ATRA d2 + polyI:C ATRA d1 + polyI:C ATRA d3 ATRA d2 ATRA d1 PolyI:C Ctrl 0
0.1 0.2 0.3 Relative expression
0.4
Cytokine and chemokine expression profile of ATRA/polyI:C-stimulated melanoma cells. WM35 and WM983A cells were treated as described in the Methods section and in Fig. 1. (a) Relative mRNA expression levels of IL-1b, IL-6, IFNb, CXCL10, CXCL9, CXCL8/IL-8, and CXCL1 in WM35 (light gray bars) or in WM983A (dark gray bars) cells. (b) Levels of secreted cytokines after ATRA and/or polyI:C treatment in WM35 (light gray bars) and WM983A (dark gray bars) cell lines. Mean±SEM values of triplicate measurements of three independent Q-PCR/ELISA experiments are shown. ATRA, all-trans retinoic acid; Ctrl, non-treated control; ELISA, enzyme-linked immunosorbant assay; IFNb, interferon b; IL, interleukin; Q-PCR, quantitative PCR.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
356
Melanoma Research 2012, Vol 22 No 5
Fig. 2 Continued
WM35
WM983A IFNβ
IL-6
(b)
35
120
30
100 80
20
pg/ml
15
60
C yI:
d3
+p ol d2
A
A
AT R
AT R
AT R
+p ol
yI:
yI: +p ol
d1 A
d3 A AT R
C
C
d3
d2
AT RA
d1
AT RA
Po l
C
AT RA
trl
C yI: +p ol
yI: +p ol
d2 A AT R
AT R
A
d1
+p ol
yI:
C
C
d3 AT RA
AT RA
AT RA
yI: Po l
C
d2
0
d1
0
C
20
trl
5
C
40
10
yI:
pg/ml
25
CXCL9
CXCL10 400
250
350 300
150
pg/ml
pg/ml
200
100
250 200 150 100
50
50
C yI: +p
ol
d3
+p AT R
A
d2 A AT R
ol
yI:
yI: ol +p
d1 A AT R
CXCL8/IL-8
C
C
d3
d1
d2
AT RA
AT RA
AT RA
Po l
yI:
trl
ol +p d3
A AT R
A AT R
C
yI:
yI: ol +p
d2
d1 A AT R
C
C
C yI: ol +p
d2
d3 AT RA
d1
AT RA
AT RA
C Po l
yI:
trl C
C
0
0
CXCL1 400
1200
350
1000
300 pg/ml
pg/ml
800 600
250 200 150
400
100 200
50
:C +p ol yI
:C A AT R
AT R
A
d2
d3
+p ol yI
:C
d1 A AT R
A
chemokines were also higher in the supernatants of melanoma cultures treated with both ATRA and polyI:C as opposed to either ATRA or polyI:C stimulation (Fig. 2b).
+p ol yI
d2
d1
d3 AT RA
AT RA
AT RA
C Po lyI :
C
:C +p
ol yI
:C d3
+p ol yI AT R
A AT R
AT R
A
d1
d2
+p o
lyI :C
d3 AT RA
d1
d2 AT RA
yI: C Po l
AT RA
trl C
trl
0
0
Interestingly and consistent with the data of Fig. 1, 1-day ATRA pretreatment with subsequent polyI:C activation led to the strongest cytokine/chemokine responses, especially
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
PRR-induced responses in melanoma cells Szabo et al. 357
Fig. 3
(a)
Percentage of migration
120 100 80 60 40
These data altogether suggest that consecutive treatment with ATRA and polyI:C strongly increases cytokine and chemokine secretion by melanoma cells, which in turn could enhance the migration of APCs to the tumor environment. Chemokine and type I IFN responses of WM35 and WM983A cells are controlled by TLR3 and MDA5 to different extents
We showed that WM35 and WM983A cells express the membrane TLR3 and the cytosolic MDA5 sensors at
WM983A C
Sp t (b)
trl Po lyI :C AT RA d AT 1 RA d2 AT AT RA RA d1 d3 + AT RA pol yI: d2 C +p AT ol RA yI: d3 C +p ol yI: C
0
Percentage of migration
CD1a+ moDC
120 100 80 60 40 20
Po lyI :C AT RA d1 AT RA d AT 2 AT RA RA d3 d1 +p AT RA ol yI: d2 C +p AT RA ol yI: d3 C +p ol yI: C
trl
0 Sp t
We have shown that the combined consecutive treatment of melanoma cells by ATRA and polyI:C led to increased expressions of TLR3 and MDA5, and induced enhanced levels of cytokines and chemokines controlled by these PRRs through the NF-kB and IRF3/IFNb signaling pathways, respectively (Figs 1 and 2). As the CXCL chemokines of interest are potent attractors of leukocytes in general, but preferentially for professional APCs, T cells, and neutrophils [13–16]), we aimed to determine a functional output of our findings. We compared the chemoattractant potential of nontreated and ATRAtreated and/or polyI:C-treated melanoma culture supernatants in migration assays using two important APCs known to be involved in skin-related antitumor immune responses, that is moMACs and DC-SIGN+/CD1a+ moDCs. We found that both moMACs (Fig. 3a) and CD1a+ moDCs (Fig. 3b) responded to melanoma supernatants. We also showed that the supernatants of both the WM35 and the WM983A melanoma cells triggered by consecutive treatments with ATRA and polyI:C were significantly more efficient in attracting APCs than those generated by an individual ATRA or polyI:C treatment (Fig. 3). In agreement with our findings showing higher secreted levels of chemokines by WM35 cells (Fig. 2b), the supernatants of activated WM35 cells induced stronger enhancement of cell migration on both APCs (Fig. 3) as compared with WM983A cells.
WM35
20
These results directly indicated that a short-term pretreatment with ATRA, followed by a 24h polyI:C activation resulted in enhanced cytokine and chemokine responses in WM35 and WM983A cells. Enhanced migration of moMACs and CD1a+ moDCs by melanoma culture supernatants after ATRA and/or polyI:C treatments
moMAC
C
those of IFNb, CXCL10, CXCL9, and CXCL8 (Fig. 2b). However, we did not find significant differences in the production of IL-6 as compared with the control treatments (Fig. 2b), and could not detect secreted levels of IL-1b, IL-10, and TNF-a (data not shown). Moreover, WM983A cells produced much higher levels of IFNb cytokine than WM35 cells upon polyI:C or ATRA + polyI:C stimulation (Fig. 2b). This effect might be attributed to the higher expression of MDA5 in these cells as shown in Fig. 1. Conversely, WM35 cultures produced higher levels of chemokines, in particular, CXCL10 and CXCL1 (Fig. 2b).
Migration of human moMACs and CD1a + moDCs stimulated by the supernatants of human melanoma in-vitro cultures. Migration of human moMACs (a) and CD1a + dendritic cells (b) was induced by the supernatants of WM35 (light gray line) and WM983A (dark gray line) cell cultures treated as described earlier. For details of the migration assay, see the Methods section. Mean±SEM data of duplicate measurements of three independent experiments are presented. ATRA, all-trans retinoic acid; Ctrl, non-treated control; moDCs, monocytederived dendritic cells; moMACs, monocyte-derived macrophage; Spt, spontaneous migration..
different levels (Fig. 1). Although polyI:C is a highly specific ligand for both TLR3 and MDA5 [20,34], TLR3 detects endocytosed polyI:C [35], and MDA5 senses polyI:C penetrating through the cell membrane by an unknown mechanism [36]. We also showed that consecutive treatments of melanoma cells in vitro by ATRA and polyI:C increased the secretion of IFNb, CXCL10, CXCL9, CXCL8, and CXCL1, which enhanced the supernatant-induced migration of professional APCs (Figs 2 and 3). Our results also indicated that the CXCL chemokines caused this migration-increasing effect, as the supernatant of WM35 cultures secreting higher levels of these proteins was a more potent inducer of APC migration (Fig. 3a and b) than the supernatant of WM983A cells. On the basis of these data, we aimed to determine the roles of TLR3 and MDA5 in mediating
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
358
Melanoma Research 2012, Vol 22 No 5
Fig. 4 WM35
WM983A
NT Ctrl siRNA
NT Ctrl siRNA
C
AT
RA
d1
AT
+p ol
yl:
d1
yl:
C
trl C
Po l
RA
pg/ml
C yl:
d1
ol +p
AT
Po
RA
lyl
C
:C
trl
C yl:
d1
ol +p
d1
d1
AT
RA
RA
AT
AT AT
C yl:
d1 AT
RA
RA
d1
d1
AT
+p
ol
RA
lyl
C
Po
ol +p
AT
:C
trl
C yl:
d1
:C
RA
lyl Po
d1
lyl
C ol
A AT R
C
+p
ol
yl:
d1
:C
d1
lyl
RA AT
C
trl Po
C yl:
d1
ol +p
∗
AT R
A
d1 RA AT
+p
d1
AT
Po
1 Ad AT R :C
RA AT
lyl
C
trl Po
C yl:
d1
ol +p
IFNβ − WM983A
∗
d1
yl:
d1
:C
trl C
yl:
+p
RA
pg/ml pg/ml
C
200 160 120 80 40 0
CXCL1 − WM983A
ol
d1
:C AT
Po
RA
lyl
C
trl
yl: C
d1
ol +p
d1 RA AT :C
A AT R RA
lyl
C
trl Po
C ol yl:
+p
MDA5 siRNA
300 250 200 150 100 50 0
IFNβ − WM35
∗
AT
AT
RA
d1
AT R
A
:C
d1
0
lyl
pg/ml
C
d1 RA AT :C
RA
lyl
C
Po trl C
C yl:
d1
ol +p
AT
1 Ad AT R
∗
20
trl
CXCL8 − WM983A
AT :C
RA AT
+p
Po
lyl
C
trl
yl: C ol
d1
∗
siRNA ctrl
CXCL1 − WM35
IFNβ − WM983A
40
C
pg/ml
Po l trl
C yl:
d1
ol +p
d1
CXCL8 − WM35
RA
RA AT
CXCL1 − WM983A
IFNβ − WM35
Po
1000 800 600 400 200 0
yl:
d1
trl C
C yl:
d1 d1
AT AT RA
lyl
d1
AT
+p
Po
ol
C
:C
trl
d1
∗
AT RA
lyl Po
C
trl
:C
∗
60 pg/ml
250 200 150 100 50 0
CXCL10 − WM983A
CXCL9 − WM983A
Nontreated CXCL1 − WM35
80
140 120 100 80 60 40 20 0
+p ol
pg/ml pg/ml pg/ml pg/ml pg/ml
100 80 60 40 20 0
AT 200 160 120 80 40 0
TLR3 siRNA CXCL8 − WM983A
RA
pg/ml pg/ml
300 250 200 150 100 50 0
C
1000 800 600 400 200 0
pg/ml
siRNA ctrl
yl:
:C
RA
lyl
C
500 400 300 200 100 0
RA :C
RA
lyl
C
Po
ol +p d1
AT RA
∗
Po
1000 800 600 400 200 0
AT
∗
AT Nontreated CXCL8 − WM35
MDA5 siRNA
CXCL9 − WM35
CXCL9 − WM983A
trl
C yl:
d1
:C
RA
lyl
C
trl
∗
Po pg/ml
500 400 300 200 100 0
600 500 400 300 200 100 0
siRNA ctrl
CXCL10 − WM35
+p ol
C
Po l
+p ol
RA AT CXCL9 − WM35
trl
pg/ml
1000 800 600 400 200 0
Nontreated 300 250 200 150 100 50 0
∗
trl
d1
CXCL10 − WM983A
d1
AT
(c)
TLR3 siRNA
C
pg/ml pg/ml
300 250 200 150 100 50 0
C
120 100 80 60 40 20 0
yl:
C yl:
C
Po l
RA
pg/ml
500 400 300 200 100 0
∗
trl
pg/ml
250 200 150 100 50 0
siRNA ctrl
CXCL10 − WM35
RA
Nontreated
(b)
C
β-Actin
RA
MDA5
β-Actin
AT
TLR3
MDA5
yl:
TLR3
yl:
(a)
The effect of TLR3 or MDA5 silencing on the cytokine secretion of ATRA-treated and/or polyI:C-treated melanoma cells. Gene-specific silencing of TLR3 and MDA5 was performed as described in the Methods section. (a) Validation of siRNA knockdown by western blot. WM35 and WM983A cells were transfected with negative control siRNA (ctrl) or with gene-targeting siRNA (siRNA), or left untreated [(nontreated controls (NT)]. Cytokine profile of TLR3 (b) and MDA5 (c) knockdown melanoma cells following ATRA and/or polyI:C activation (see the Methods section). White and black bars represent nontreated and siRNA controls, respectively. Gray bars show the effect of TLR3 (b) or MDA5 (c) silencing on the secreted levels of CXCL10, CXCL9, CXCL8, CXCL1, and IFNb, as measured by ELISA. Mean±SEM values of duplicates of three independent experiments are presented. ATRA, all-trans retinoic acid; Ctrl, non-activated control; ELISA, enzyme-linked immunosorbant assay; IFNb, interferon b; TLR3, toll-like receptor 3. *P < 0.05.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
PRR-induced responses in melanoma cells Szabo et al.
359
Fig. 5
20 0
Po
d1 AT RA
A AT R 80 ∗
60 40 20
Percentageof migration (moMAC)
WM35
100
0
WM983A
100 80 60
∗
40 20
siRNA ctrl
20
ol
yI:
C
d1 AT
AT RA
RA
d1
d1
+p
Po
+p
AT RA
yI: ol
:C
0
C
d1 AT RA
Po
lyI
:C
trl
0
40
lyI
20
60
trl
40
80
C
60
100
Sp t
Percentage of migration (CD1a+ moDC)
80
C
yI:
WM983A
100
WM35 Percentage of migration (moMAC)
100
ol
MDA5 siRNA
WM35
Sp t 80 60 40 20 0
WM983A
100 80 60 40 20
ol yI: C
d1
Po lyI :C
trl C
Sp t
AT RA
d1 RA AT
AT RA
d1
+p
ol yI: C
d1 AT RA
Po lyI :C
C trl
Sp t
0
+p
Percentage of migration (CD1a+ moDC) Percentage of migration (moMAC)
(d)
+p d1
AT RA
d1 RA AT Nontreated
(c)
C
d1
:C
AT RA
+p
Po
lyI
C trl
Sp t
C yI: ol
AT RA
d1
:C lyI Po
C
trl
0
Sp t
Percentage of migration (moMAC)
(b)
yI: C
Sp t
d1
+p
AT R
ol
A
yI: C
d1
lyI :C
trl
Po
C
Sp t
0
∗
40
ol
20
60
d1
40
+p
60
80
AT RA
∗
WM983A
100
lyI :C
80
TLR3 siRNA
trl
100
Percentage of migration (CD1a+ moDC)
Percentage of migration (CD1a+ moDC)
siRNA ctrl
WM35
C
Nontreated
(a)
Influence of TLR3 or MDA5 knockdown on the migration of human monocyte-derived APCs induced by tumor supernatants of ATRA-treated and/or polyI:C-treated melanoma cell cultures. Migration profiles of human moMACs and CD1a + moDCs stimulated by tumor supernatants of WM35 or WM983A treated as described earlier. (a, b) The effect of TLR3 silencing on the migration of CD1a + moDCs (a) or moMACs (b). (b, d) Migration of CD1a + moDCs (c) and moMACs (d) after transfection by MDA5-specific siRNA. In each panel, white and black bars show nontreated and negative control siRNA-treated controls, respectively. Gray bars represent TLR3-specific or MDA5-specific siRNA treatment. Mean±SEM values of triplicate measurements of two independent experiments are shown. APCs, antigen-presenting cells; ATRA, all-trans retinoic acid; Ctrl, non-activated control; moDCs, monocyte-derived dendritic cells; moMACs, monocyte-derived macrophage; Spt, spontaneous migration; TLR3, toll-like receptor 3. *P < 0.05.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
360 Melanoma Research 2012, Vol 22 No 5
cytokine responses by analyzing the activity of the NF-kB and IRF3 pathways that control the production of CXCL chemokines and IFNb, respectively. As chemokines and IFNb levels peaked after a 1-day ATRA pretreatment, followed by a 24h polyI:C stimulation (Fig. 2b), we used this setup and the siRNA technology to address this question. The immunoblots shown in Fig. 4a show the efficacy of the PRR-specific siRNAs in downregulating TLR3 and MDA5 in both cell lines. It also shows that silencing of TLR3 completely abrogated CXCL10, CXCL9, CXCL8, and CXCL1 chemokine secretion (Fig. 4b), and caused a moderate but significant decrease in IFNb production in both cell lines (Fig. 4b). Conversely, knockdown of MDA5 did not affect the production of CXCL chemokines, but significantly reduced IFNb secretion (Fig. 4c). This suggests that the production of chemokines is controlled by endosomal TLR3 in WM35 and WM983A melanoma cells, whereas the type I IFN response is controlled by both TLR3 and MDA5. Our findings are in agreement with previous reports showing the disparate regulation of NF-kB-related and IRF3–IFNb-related signaling in immune cells upon polyI:C stimulation [37,38]. TLR3 is responsible for the enhanced migration of APCs induced by the supernatants of ATRA + polyI:C-activated melanoma cultures
We showed that MDA5 is not required for polyI:Cinduced secretion of CXCL10, CXCL9, CXCL8, and CXCL1 in WM35 and WM983A cells (Fig. 4c), but TLR3 plays a major role in this process (Fig. 4b). As these CXCL chemokines are crucial for the in vivo induction and control of the directed migration of several leukocytes, we checked the results of TLR3 and MDA5 silencing on the migration-inducing capacity of ATRA + polyI:C-activated melanoma culture supernatants. We found that gene-specific knockdown of TLR3 strongly reduced the migration-enhancing effect of the supernatants of both cell lines, resulting in a significant decrease in the percentage of migrated CD1a + moDCs (Fig. 5a) and moMACs (Fig. 5b). Silencing of MDA5, however, did not affect the migration-inducing effect of tumor supernatants (Fig. 5c and d). These results show that TLR3, but not MDA5, plays an important functional role in controlling the secretion of chemoattractant peptides in polyI:C-activated melanoma cells.
Discussion In the last three decades, melanoma has become an increasing threat within the circle of skin malignancies [1]. Besides standard chemotherapy, activation, and mobilization of professional APCs to increase protective immune responses, cytokine and chemokine production as well as effector T lymphocyte generation has emerged as a potential treatment modality [2]. For example, an
impressive novel strategy is based on the induction and mobilization of professional APCs and CTLs [1,5]. Furthermore, both ATRA and polyI:C are promising candidates for future melanoma therapies [24–32]. In this study, we determined the effects of consecutive ATRA and polyI:C treatments in in-vitro melanoma cell cultures. Our results showed that short-term, consecutive combined treatments with ATRA and polyI:C increased the expression of TLR3 and MDA5 in WM35 and WM983A melanoma cells at both transcriptional and protein levels, and even a 24h presensitization of melanoma cells by ATRA could significantly enhance the activating capability of polyI:C to induce type I IFN and chemokine secretion. A 24h pretreatment with 1 mmol/l ATRA, followed by polyI:C activation also enhanced the mRNA and protein expression levels of IFNb, CXCL10, CXCL9, CXCL8, and CXCL1 in both melanoma cells, and this effect was attributed to the polyI:C-specific receptors TLR3 and MDA5. However, gene silencing experiments showed that membrane TLR3 and cytosolic MDA5 contribute toward these effects to different extents as TLR3 was found to be exclusively responsible for controlling CXCL chemokine secretion, whereas both TLR3 and MDA5 were involved in the polyI:C-induced IFNb response in melanoma. These results also confirmed that human moMACs and CD1a+ moDC can be mobilized and attracted efficiently by the released cytokines and chemokines of ATRA + polyI: C-treated melanoma culture supernatants, and this process was solely controlled by the specific activation of TLR3. To the best of our knowledge, we have shown for the first time that short-term ATRA pretreatment of melanoma cells can enhance the polyI:C-dependent activation of both TLR3 and MDA5 in melanoma cells, as was found at the level of cytokine and chemokine genes and proteins. Moreover, this is the first report to show that the control of polyI:C-induced chemokine responses in melanoma cells is uniquely based on the activation of TLR3, whereas both TLR3 and MDA5 play a role in the regulation of IFNb secretion. As IFNb is not only an antiviral agent but has also been shown to act as a strong anticancer factor [11,12,39], we propose that this novel treatment option – by means of the enhancement of type I IFN and chemokine responses and by the potent mobilization of APCs – may have a major impact on the efficacy of future melanoma therapies.
Acknowledgements The authors thank Ro´za Za´ka´ny (Department of Anatomy, Histology and Embryology, UD-MHSC) for reviewing the manuscript. The excellent technical assistance of Erzse´bet Nagy is highly appreciated. ´ MOP-4.2.1/B-09/1/ This work was supported by the TA KONV-2010–0007 and OTKA NK101538 grants.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
PRR-induced responses in melanoma cells Szabo et al.
Conflicts of interest
21
There are no conflicts of interest.
References 1
2 3 4 5 6 7
8 9
10 11 12
13
14 15
16
17
18
19
20
Sharma P, Wagner K, Wolchok JD, Allison JP. Novel cancer immunotherapy agents with survival benefit: recent successes and next steps. Nat Rev Cancer 2011; 11:805–812. Eggermont AM, Robert C. New drugs in melanoma: it’s a whole new world. Eur J Cancer 2011; 47:2150–2157. Sullivan RJ, Atkins MB. Cytokine therapy in melanoma. J Cutan Pathol 2010; 37 (Suppl 1):S60–S67. Aamdal S. Current approaches to adjuvant therapy of melanoma. Eur J Cancer 2011; 47 (Suppl 3):S336–S337. Roman RA. Immunotherapy for advanced melanoma. Clin J Oncol Nurs 2011; 15:E58–E65. Nestle FO, Di Meglio P, Qin JZ, Nickoloff BJ. Skin immune sentinels in health and disease. Nat Rev Immunol 2009; 9:679–691. Merad M, Ginhoux F, Collin M. Origin, homeostasis and function of Langerhans cells and other langerin-expressing dendritic cells. Nat Rev Immunol 2008; 8:935–947. Kawai T, Akira S. Toll-like receptors and their crosstalk with other innate receptors in infection and immunity. Immunity 2011; 34:637–650. Benko S, Magyarics Z, Szabo A, Rajnavolgyi E. Dendritic cell subtypes as primary targets of vaccines: the emerging role and cross-talk of pattern recognition receptors. Biol Chem 2008; 389:469–485. Kawai T, Akira S. Toll-like receptor and RIG-I-like receptor signaling. Ann NY Acad Sci 2008; 1143:1–20. Gonzalez-Navajas JM, Lee J, David M, Raz E. Immunomodulatory functions of type I interferons. Nat Rev Immunol 2012; 12:125–135. Caraglia M, Marra M, Tagliaferri P, Lamberts SW, Zappavigna S, Misso G, et al. Emerging strategies to strengthen the anti-tumour activity of type I interferons: overcoming survival pathways. Curr Cancer Drug Targets 2009; 9:690–704. Albanesi C, Scarponi C, Giustizieri ML, Girolomoni G. Keratinocytes in inflammatory skin diseases. Curr Drug Targets Inflamm Allergy 2005; 4:329–334. Lacotte S, Brun S, Muller S, Dumortier H. CXCR3, inflammation, and autoimmune diseases. Ann NY Acad Sci 2009; 1173:310–317. Neville LF, Mathiak G, Bagasra O. The immunobiology of interferon-gamma inducible protein 10 kD (IP-10): a novel, pleiotropic member of the C-X-C chemokine superfamily. Cytokine Growth Factor Rev 1997; 8:207–219. Liu M, Guo S, Hibbert JM, Jain V, Singh N, Wilson NO, et al. CXCL10/IP-10 in infectious diseases pathogenesis and potential therapeutic implications. Cytokine Growth Factor Rev 2011; 22:121–130. Amatschek S, Lucas R, Eger A, Pflueger M, Hundsberger H, Knoll C, et al. CXCL9 induces chemotaxis, chemorepulsion and endothelial barrier disruption through CXCR3-mediated activation of melanoma cells. Br J Cancer 2011; 104:469–479. Singh S, Singh AP, Sharma B, Owen LB, Singh RK. CXCL8 and its cognate receptors in melanoma progression and metastasis. Future Oncol 2010; 6:111–116. Antonicelli F, Lorin J, Kurdykowski S, Gangloff SC, Le Naour R, Sallenave JM, et al. CXCL10 reduces melanoma proliferation and invasiveness in vitro and in vivo. Br J Dermatol 2011; 164:720–728. Alexopoulou L, Holt AC, Medzhitov R, Flavell RA. Recognition of double-stranded RNA and activation of NF-kappaB by toll-like receptor 3. Nature 2001; 413:732–738.
22
23 24
25
26
27 28
29
30
31
32
33
34
35 36
37
38
39
361
Yu M, Levine SJ. Toll-like receptor, RIG-I-like receptors and the NLRP3 inflammasome: key modulators of innate immune responses to doublestranded RNA viruses. Cytokine Growth Factor Rev 2011; 22:63–72. Mathe G, Kamel M, Dezfulian M, Halle-Pannenko O, Bourut C. An experimental screening for ‘systemic adjuvants of immunity’ applicable in cancer immunotherapy. Cancer Res 1973; 33:1987–1997. Cheng YS, Xu F. Anticancer function of polyinosinic-polycytidylic acid. Cancer Biol Ther 2011; 10:1219–1223. Besch R, Poeck H, Hohenauer T, Senft D, Hacker G, Berking C, et al. Proapoptotic signaling induced by RIG-I and MDA-5 results in type I interferon-independent apoptosis in human melanoma cells. J Clin Invest 2009; 119:2399–2411. Salaun B, Lebecque S, Matikainen S, Rimoldi D, Romero P. Toll-like receptor 3 expressed by melanoma cells as a target for therapy? Clin Cancer Res 2007; 13:4565–4574. Fujimura T, Nakagawa S, Ohtani T, Ito Y, Aiba S. Inhibitory effect of the polyinosinic-polycytidylic acid/cationic liposome on the progression of murine B16F10 melanoma. Eur J Immunol 2006; 36:3371–3380. Jiang Q, Wei H, Tian Z. Poly I:C enhances cycloheximide-induced apoptosis of tumor cells through TLR3 pathway. BMC Cancer 2008; 8:12–19. Salem ML, Kadima AN, Cole DJ, Gillanders WE. Defining the antigenspecific T-cell response to vaccination and poly(I:C)/TLR3 signaling: evidence of enhanced primary and memory CD8 T-cell responses and antitumor immunity. J Immunother 2005; 28:220–228. Kast RE. Potential for all-trans retinoic acid (tretinoin) to enhance interferonalpha treatment response in chronic myelogenous leukemia, melanoma, myeloma and renal cell carcinoma. Cancer Biol Ther 2008; 7:1515–1519. Desai SH, Boskovic G, Eastham L, Dawson M, Niles RM. Effect of receptorselective retinoids on growth and differentiation pathways in mouse melanoma cells. Biochem Pharmacol 2000; 59:1265–1275. Zhang H, Satyamoorthy K, Herlyn M, Rosdahl I. All-trans retinoic acid (atRA) differentially induces apoptosis in matched primary and metastatic melanoma cells – a speculation on damage effect of atRA via mitochondrial dysfunction and cell cycle redistribution. Carcinogenesis 2003; 24:185–191. Baroni A, Paoletti I, Silvestri I, Buommino E, Carriero MV. Early vitronectin receptor downregulation in a melanoma cell line during all-trans retinoic acid-induced apoptosis. Br J Dermatol 2003; 148:424–433. Gogolak P, Rethi B, Szatmari I, Lanyi A, Dezso B, Nagy L, et al. Differentiation of CD1a – and CD1a + monocyte-derived dendritic cells is biased by lipid environment and PPARgamma. Blood 2007; 109:643–652. Kato H, Takeuchi O, Sato S, Yoneyama M, Yamamoto M, Matsui K, et al. Differential roles of MDA5 and RIG-I helicases in the recognition of RNA viruses. Nature 2006; 441:101–105. Matsumoto M, Seya T. TLR3: interferon induction by double-stranded RNA including poly(I:C). Adv Drug Deliv Rev 2008; 60:805–812. Wang Y, Cella M, Gilfillan S, Colonna M. Cutting edge: polyinosinic:polycytidylic acid boosts the generation of memory CD8 T cells through melanoma differentiation-associated protein 5 expressed in stromal cells. J Immunol 2010; 184:2751–2755. McCartney S, Vermi W, Gilfillan S, Cella M, Murphy TL, Schreiber RD, et al. Distinct and complementary functions of MDA5 and TLR3 in poly(I:C)mediated activation of mouse NK cells. J Exp Med 2009; 206:2967–2976. Szabo A, Bene K, Gogola´k P, Re´thi B, La´nyi A, Jankovich I, et al. RLR-mediated production of interferon-b by a human dendritic cell subset and its role in virus-specific immunity. J Leukoc Biol 2012; 92:159–169. Xiao HB, Zhou WY, Chen XF, Mei J, Lv ZW, Ding FB, et al. Interferon-beta efficiently inhibited endothelial progenitor cell-induced tumor angiogenesis. Gene Ther 2011. doi: 10.1038/gt.2011.171.
Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
