Upload
dinhdan
View
215
Download
0
Embed Size (px)
Citation preview
1
West Nile virus non-coding subgenomic RNA contributes to viral evasion of 1
type I interferon-mediated antiviral response 2
Running Title: sfRNA interferes with anti-viral response 3
4
Andrea Schuessler1#, Anneke Funk1#£, Helen M. Lazear2, Daphne A. Cooper3 , Shessy Torres1, 5
Stephane Daffis2§, Babal Kant Jha4, Yutaro Kumagai5, Osamu Takeuchi5, Paul Hertzog6 , 6
Robert Silverman4 , Shizuo Akira5, David J. Barton3, Michael S. Diamond2, Alexander A. 7
Khromykh1* 8
9
1Australian Infectious Disease Research Centre, School of Chemistry and Molecular 10
Biosciences, University of Queensland, Brisbane, QLD 4072, Australia 11
2Departments of Medicine, Molecular Microbiology, Pathology & Immunology, Washington 12
University School of Medicine, Saint Louis, MO, 63110 USA 13
3Department of Microbiology, University of Colorado School of Medicine, Aurora, CO, 14
80045, USA 15
4 Lerner Research Institute, Cleveland, Ohio, 44195, USA 16
5Department of Host Defence, University of Osaka, Osaka, Japan 17
6Monash Institute of Medical Research, Clayton, 3168, Victoria, Australia 18
19
# these authors contributed equally 20
£ Present address: EMBO, Meyerhofstr. 1, 69115 Heidelberg 21
§ Present address: Gilead Sciences, Inc, Foster City, CA 94404 22
*corresponding author: Alexander A. Khromykh, email: [email protected] 23
word count abstract: 180 24
word count text (excluding the references, table footnotes, and figure legends): 4,60925
Copyright © 2012, American Society for Microbiology. All Rights Reserved.J. Virol. doi:10.1128/JVI.00207-12 JVI Accepts, published online ahead of print on 29 February 2012
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
2
ABSTRACT 26
We previously showed that a flavivirus non-coding subgenomic RNA (sfRNA) is 27
required for viral pathogenicity, as a mutant West Nile virus (WNV) deficient in sfRNA 28
production replicated poorly in wild type mice. To investigate the possible immunomodulatory 29
or immune evasive functions of sfRNA, we utilized mice and cells deficient in elements of the 30
type I interferon (IFN) response. Replication of the sfRNA mutant WNV was rescued in mice 31
and cells lacking interferon regulatory factors (IRF)-3 and -7 and in mice lacking the type I 32
interferon-αβ receptor (IFNAR) suggesting a contribution for sfRNA in overcoming the 33
antiviral response mediated by type I IFN. This was confirmed by demonstrating rescue of 34
mutant virus replication in the presence of IFNAR neutralizing antibodies, greater sensitivity 35
of mutant virus replication to IFN-α pre-treatment, partial rescue of its infectivity in cells 36
deficient in RNase L, and direct effects of transfected sfRNA on rescuing replication of 37
unrelated Semliki Forest virus in cells pre-treated with IFN-α. The results define a novel 38
function of sfRNA in flavivirus pathogenesis via contributing to viral evasion of type I 39
interferon response. 40
41
42
43
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
3
INTRODUCTION 44
West Nile virus (WNV) is a member of the Flavivirus genus of the Flaviviridae family 45
of RNA viruses, and is closely related to a number of human pathogens of global concern 46
including dengue (DENV), yellow fever (YFV), tick-borne encephalitis (TBEV), and Japanese 47
encephalitis (JEV) viruses. Many flaviviruses cause fatal disease in humans and outbreaks 48
affect 50 to 100 million people every year (20, 36). Since 1999, highly pathogenic North 49
American strains of WNVhave caused more than 30,000 clinical cases of meningitis, 50
encephalitis, and acute flaccid paralysis in the United States alone. In comparison, the 51
Australian strains of WNV, which are referred to as Kunjin virus (WNVKUN), are closely 52
related (~97% homology at the amino acid level) but do not cause disease in 53
immunocompetent adult animals and humans (22). 54
The flavivirus genome is a single-stranded, positive polarity RNA of ~11 kb. It 55
contains one open reading frame flanked by 5’ and 3’ untranslated regions (UTRs), and 56
encodes 10 viral proteins that are required for the complete viral life cycle (30-34, 58, 59). The 57
UTRs play essential functions in the initiation of RNA replication, translation, and genome 58
packaging (40). In addition to full-length genomic RNA (gRNA), an abundant RNA species of 59
about 0.5 kb derived from the 3’UTR of gRNA was previously detected in flavivirus-infected 60
cells (29, 44, 49, 57) and termed subgenomic flavivirus RNA (sfRNA). Recent studies 61
demonstrated that the sfRNA of WNVKUN and YFV is generated as a product of degradation 62
by a host enzyme, presumably the 5’-3’ exoribonuclease XRN1 (44, 51). XRN1-mediated 63
degradation of gRNA likely stalls due to the rigid and conserved secondary and tertiary RNA 64
structures in the 5’end of the 3’UTR (18, 51). Thus, incomplete degradation of WNVKUN RNA 65
results in a 525 nucleotide (nt) RNA remnant that forms the sfRNA. The sfRNA contributes to 66
virus-induced cytopathic effect in cell culture and to virulence in highly sensitive to flavivirus 67
infections weanling mice (44), although the mechanism(s) that explain these outcomes remain 68
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
4
unknown. Because of the requirement of sfRNA for virulence in mice but not for replication in 69
BHK-21 or Vero cells that lack intact cell-intrinsic antiviral immune pathways, we 70
hypothesized that sfRNA modulates the host antiviral response. 71
Viral infection of host cells results in the induction of cell-intrinsic and cell-extrinsic 72
antiviral responses that limit replication and spread. Pathogen recognition receptors (PRRs), 73
including the cytoplasmic receptors retinoid acid-inducible gene I (RIG-I) and melanoma 74
differentiation-associated gene-5 (MDA5) or membrane-bound Toll-like receptors (TLR3, 7 or 75
8) serve as initial sensors of pathogen-associated molecular patterns (PAMPs) after RNA virus 76
infection and trigger signal transduction cascades that induce the expression of genes with 77
specific inhibitory functions. For RNA viruses, double-stranded RNA (dsRNA) intermediates 78
of gRNA replication are believed to be the primary PAMPs. Activation of PRRs results in 79
signaling through distinct adaptor molecules. RIG-I and MDA5 signal through interferon-80
β promoter stimulator 1 (IPS-1), whereas TLR3 and TLR7/8 signal through Trif or MyD88, 81
respectively. Ultimately, these pathways result in phosphorylation and activation of 82
transcription factors (e.g. interferon regulatory factor 3 [IRF-3], and IRF-7), which together 83
with nuclear factor (NF)-кB and ATF-2/c-jun induce transcription of antiviral cytokines such 84
as IFN-α4 (39) and IFN-β (54). Secreted type I IFN binds to the IFN-α/β receptor (IFNAR) in 85
an autocrine and paracrine manner and activates the Janus kinase/signal transducers and 86
activators of transcription (JAK/STAT) signaling cascade. This leads to formation of the 87
ISGF3 complex (STAT1, STAT2, and IRF9/p48), which translocates into the nucleus and 88
induces expression of several hundred IFN-stimulated genes (ISGs), many of which likely 89
have antiviral functions. A number of ISGs, including RNase L, PKR, IFIT-1, IFIT-2, ISG20, 90
IFITM3, viperin and others are believed to possess antiviral activity against flaviviruses (4, 14, 91
25, 47, 49). 92
To investigate whether sfRNA modulates virus-host interactions, we compared 93
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
5
replication of wild type (wt) WNVKUN virus to a mutant virus incapable of producing sfRNA 94
utilizing cells and mice deficient in various components of the innate antiviral response. Using 95
this approach, we show that sfRNA contributes to viral evasion of the type I IFN-mediated 96
antiviral response. 97
98
99
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
6
MATERIALS AND METHODS 100
Viruses and cells. Wild type (wt) WNVKUN full-length cDNA clone FLSDX (pro)_HDVr 101
(designated FLSDX) and sfRNA-deficient virus FL-IRAΔCS3 were described previously (30, 102
44). For virus reconstitution, viral RNA was in vitro transcribed from the corresponding full-103
length cDNA clones, electroporated into BHK-21 cells and subsequently passaged once on 104
Vero76 cells to grow a high titer stock. Passage 2 virus was titered by standard plaque assay 105
on BHK-21 cells and used for experiments. Semliki Forest virus (SFV) was grown and titered 106
on Vero76 cells. WNV RNA for in vitro RNase L assays was transcribed from C20DXrep, a 107
cDNA construct containing the FLSDX sequence, but lacking the structural genes C, PrM and 108
E (27). 109
For WNVKUN infection experiments, all murine embryonic fibroblasts (MEFs) with the 110
exception of RNase L-/- MEFs were low passage primary cultures (n = 2-10) generated from 111
wt and deficient mice. RNase L-/- and corresponding control MEFs were immortalized after 112
transformation with SV40 T antigen (61). IRF-3-/- x IRF-7-/- MEFs for IFN sensitivity 113
experiments and control WT MEFs for sfRNA electroporation and SFV rescue experiments 114
were immortalized spontaneously by iterative passaging. 115
WNVKUN infection, growth kinetics and plaque assays. MEFs were infected at different 116
multiplicities of infection (MOI) for 2 h at 37°C, washed three times with PBS, and incubated 117
with DMEM containing 5% FBS. For viral growth kinetics, cell culture supernatant was 118
harvested at the indicated times to determine virus titers by standard plaque assay on BHK-21 119
cells (44). 120
RNA isolation and Northern blotting. RNA from infected cells was isolated with Trizol 121
reagent (Invitrogen) following the manufacturer’s recommendations and Northern blots were 122
performed using 32P-labelled cDNA probes specific for WNVKUN 3’UTR and SFV nsP4 gene 123
as described previously (44). 124
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
7
In vitro sfRNA transcription. The template for in vitro transcription was generated by PCR 125
from plasmids pBS3’XX (containing T7 promoter and the last 813 nucleotides of viral cDNA 126
including part of WNVKUN NS5 and the entire 3’UTR) and pBS3’XX-IRAΔCS3 (as above, 127
but containing IRAΔCS3 mutations in the 3’UTR), respectively (44). The PCR product was 128
purified with Wizard SV Gel and PCR Clean-Up System (Promega) and 1 µg was used for in 129
vitro transcription with Guanosine 5′-monophosphate and T7 polymerase according to the 130
manufacturer’s instructions (Promega). 131
RNA electroporation, IFN treatment and SFV infection. Wild type MEFs were trypsinized 132
and resuspended at 2 x 106 cells/ml in OptiMEM (Invitrogen). 400µl cell suspension was 133
mixed with ~ 10 µg of in vitro transcribed RNA and electroporated in a 4 mm gap cuvette with 134
a single pulse at 400V for 5ms using an ECM 830 square wave electroporator (BTX, Harvard 135
apparatus). Cells were seeded in 6-well plates and left to attach for 1 to 2 hours before 136
treatment with 1000 IU/ml mouse IFN-α (Hycult Biotechnology) for 8 hours. The IFN 137
solution was removed and cells were infected with SFV at an MOI of 1 for 1 hour. At 12 hours 138
post-infection (hpi), cells were lysed with TriReagent (Sigma) for RNA extraction and 139
supernatants were harvested for plaque assay on Vero76 cells. 140
Virulence in mice. All animal studies were carried out in strict accordance with the 141
recommendations in the Guide for the Care and Use of Laboratory Animals of the National 142
Institutes of Health USA. The protocols were approved by the Institutional Animal Care and 143
Use Committee at the Washington University School of Medicine, Assurance Number: 144
A3381-01). All inoculation and experimental manipulation was performed under anesthesia 145
that was induced and maintained with ketamine hydrochloride and xylazine, and all efforts 146
were made to minimize suffering. 17 to 20 day old wild type C57BL/6 mice were inoculated 147
subcutaneously into the footpad with 103 PFU of WNVKUN FLSDX or FL-IRAΔCS3 in HBSS 148
supplemented with 1% FBS, monitored for the signs of illness, and sacrificed when signs of 149
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
8
encephalitis became apparent. Mice were bled at days 2 and 3 after infection and viral 150
genomic RNA was determined by real-time RT-PCR, as described previously (47). 151
8 to 12 week old IRF-3-/- x IRF-7-/- and IFNAR-/- mice (13) both on C57BL/6 genetic 152
background, were inoculated via footpad with various doses (101, 102, and 103 PFU) of wt 153
(FLSDX) or sfRNA-deficient virus (FL-IRAΔCS3). 154
In vitro RNase L assays. 150nM of radiolabeled in vitro transcribed WNV RNA 155
(C20DXrep), RNase L-resistant poliovirus (PV) RNA and RNase L-sensitive PV RNA G5761A 156
were incubated for specified times with 20 nM RNase L and 20 nM 2-5A (negative control 157
was incubated without 2-5A). For inhibition experiments, radiolabeled in vitro transcribed 158
WNV RNA was incubated for 60 minutes in reactions containing 20 nM RNase L and 20 nM 159
2-5A or 20 nM RNase L without 2-5A. 500, 1000, and 2000 nM sfRNA, PV2122 ciRNA, and 160
PV2121 RNA were included in the reactions. All reactions were terminated by addition of SDS 161
buffer, phenol-chloroform extracted, ethanol precipitated with tRNA carrier, and fractionated 162
by electrophoresis in agarose. The integrity of RNAs was analyzed by ethidium bromide 163
staining and phosphorimaging. 164
Statistical analyses. Statistical significance of IFN sensitivity experiments was analysed by 165
regression analysis using R program (45). All other data were analyzed using Prism software 166
(GraphPad Prism). A 2-way ANOVA was used to analyze differences in viral burden in mice. 167
Kaplan-Meier survival curves were analyzed by the log rank test. Fold decrease of SFV titers 168
after electroporation and IFN treatment was analysed by unpaired T test. 169
170
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
9
RESULTS 171
Antiviral responses mediated by IRF-3 and IRF-7 restrict replication of sfRNA-172
deficient virus in MEFs. Our previous studies have shown that sfRNA is required for 173
virulence in vivo, as an sfRNA-deficient mutant virus (FL-IRAΔCS3) failed to cause mortality 174
in the wild type (wt) weanling Swiss mice after intraperitoneal infection even at high virus 175
doses (104 PFU) (17, 44). In contrast, sfRNA-deficient mutants replicated like wild type virus 176
(FLSDX) in cells (e.g., BHK21-15 and Vero) lacking intact cell-intrinsic antiviral response 177
pathways. Thus, we hypothesized that sfRNA modulates the host antiviral response. To 178
investigate this, we performed viral growth analyses in MEFs deficient in various components 179
of the type I interferon response. IRF-3 and IRF-7 are key transcriptional regulators of IFN 180
induction and ISG expression in response to viral infections (1, 24, 48). Using IRF-3-/- x IRF-181
7-/- MEFs, we investigated whether IRF-3 and IRF-7 are required for induction of the antiviral 182
response that limits replication of sfRNA-deficient virus. While the sfRNA-deficient mutant 183
FL-IRAΔCS3 replicated poorly in wt MEFs (Figs. 1A and B, upper panels), infection at an 184
MOI of 1 of IRF-3-/- x IRF-7-/- cells with FL-IRAΔCS3 virus resulted in a substantial increase 185
in RNA replication and virus production, particularly during the first two days after infection 186
(Figs. 1A and B, lower panels). Rescue of viral RNA replication and virus production in the 187
IRF-3-/- x IRF-7-/- MEFs was rapid and approached levels of the wt virus as early as 12 to 24 188
hpi. A similar rescue was observed in IRF-3-/- x IRF-7-/- MEFs when lower levels (MOI of 0.1 189
or 0.01) of sfRNA-deficient FL-IRAΔCS3 virus was used (Fig. 1C). Again, replication of 190
mutant virus at lower MOIs was as efficient as that of the wt virus. These results demonstrate 191
that the IRAΔCS3 mutation did not affect virus replication in the absence of IRF-3 and IRF-7, 192
thus confirming our prior data in BHK-21 cells, another cell line known to be deficient in IFN 193
response (17), and establish that the mutations that abolish sfRNA production do not 194
inherently attenuate virus replication. The results also suggest that IRF-3/IRF-7-dependent 195
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
10
transcription is a primary factor limiting replication of viruses deficient in sfRNA production 196
and imply a role for sfRNA in counteracting this antiviral pathway. 197
sfRNA contributes to higher viral resistance to type I IFN in MEFs. To begin to test 198
whether sfRNA contributes to viral evasion of the IFN response, we compared the sensitivity 199
of wt and mutant viruses to IFN-α pre-treatment in IRF-3-/- x IRF-7-/- MEFs. Replication of wt 200
and sfRNA-deficient viruses is equivalent in these cells (see Fig. 1A) and they produce little, 201
if any type I IFN after WNV infection (13) so the antiviral effects will reflect only 202
exogenously administered IFN. Cells were pre-treated with increasing concentrations of IFN-α 203
for 8 hours, infected with wt or sfRNA-deficient viruses at an MOI of 1, and incubated for 48 204
hours before harvesting viral RNA from cells and titering extracellular virus. Although 205
starting at slightly lower levels, viral RNA amounts for sfRNA-deficient virus were reduced to 206
a greater extent than those for wt virus in cells treated with increasing concentrations of IFN-207
α. Whlie wt virus RNA could still be detected in cells treated with 1000 IU/ml IFN-α, RNA of 208
the sfRNA-deficient virus was undetectable after treatment with as little as 100 IU/ml (Fig. 209
2A). To further quantify this reduction, viral titers were measured by plaque assay and 210
showed a significantly larger reduction in virus yield for the sfRNA mutant virus, ranging 211
from ~10 to 1000-fold reduction for IFN-α concentrations from 1 to 1000 IU/ml (P < 0.01), 212
respectively (Fig. 2 B). To confirm the role of type I IFN as the primary factor limiting 213
replication of the mutant virus, wt MEFs were treated with different concentrations of a 214
neutralizing IFNAR-1 monoclonal antibody (MAR1-5A3), which binds to the IFNAR-1 215
subunit and prevents IFN signaling (50). Treatment of wt MEFs with MAR1-5A3 but not the 216
isotype control antibody rescued replication of sfRNA-deficient virus in wt MEFs and also 217
increased wt virus replication (Fig. 2C). These experiments indicate that the type I IFN 218
response downstream of IFNAR signaling has a primary role in restricting replication of 219
sfRNA-deficient virus in MEFs. 220
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
11
Virulence of sfRNA-deficient virus is partially rescued in IRF-3-/- x IRF-7-/- and 221
IFNAR-/- mice. To investigate whether the findings in MEFs translated into an effect on 222
virulence of sfRNA-deficient virus in mice, wt and IRF-3-/- x IRF-7-/- mice were initially 223
infected via footpad injection with 103 PFU of FLSDX and FL-IRAΔCS3 viruses. As adult wt 224
mice are not susceptible to infection with a wt WNVKUN virus (9), weanling (17 to 20 day-old) 225
wt mice were used. In contrast, adult IRF-3-/- x IRF-7-/- mice were used for infections as they 226
are highly susceptible to infection with WNVKUN virus (9). All wt weanling mice succumbed 227
to infection with wt virus by day 15 after infection (n = 16), whereas 50% of animals survived 228
infection with sfRNA-deficient virus (Fig. 3A, P < 0.002, n = 18). Virulence of the sfRNA-229
deficient virus was largely rescued in IRF-3-/-x IRF-7-/- mice and approached that of the wt 230
virus with only 20% of animals surviving the infection (Fig. 3B, P = 0.02, n = 10). Our 231
previous studies had indicated that wt virus replicated more efficiently in IFNAR-/- mice than 232
in IRF-3-/- x IRF-7-/- mice (9). To examine whether virulence of sfRNA-deficient virus was 233
rescued more efficiently in IFNAR-/- mice, we infected them with the same dose (103 PFU) of 234
wt and sfRNA-deficient virus. All animals died by day 8 after infection with mutant virus, 235
whereas wt virus killed all IFNAR-/- mice by day 5 after infection (Fig. 3C). Quantitative RT-236
PCR analysis of viral RNA in serum showed a 137-fold decrease in serum virus load at 3dpi 237
for wt mice infected with the sfRNA-deficient virus compared to wt virus, but only a 6-fold 238
decrease for the mutant virus in IFNAR-/- mice (Fig. 3F and G). As IFNAR-/- mice are highly 239
sensitive to infection, it may be difficult to discern phenotypes with a high (103 PFU) dose of 240
input virus. To address this, we also infected IFNAR-/- mice with 10- and 100-fold lower 241
amounts of virus. However, lower viral doses resulted in similar kinetics of pathogenesis (Fig. 242
3D and E) and viral burden in serum (Fig. 3H and I) thus demonstrating that rescue of 243
virulence of sfRNA-deficient virus in the absence of IFNAR-mediated anti-viral response was 244
independent of viral dose. For all viral doses tested, wt virus was more virulent, suggesting 245
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
12
that a type I IFN-independent pathway also contributed to restricting replication of the sfRNA-246
deficient virus in mice. 247
sfRNA partially rescues replication of Semliki Forest virus in IFN-treated MEFs. To 248
examine whether sfRNA can inhibit the IFN-induced antiviral response directly in the absence 249
of active WNVKUN replication, an in vitro transcribed 5’-guanosinemonophosphate (5’GMP) 250
RNA encompassing the last 813 nucleotides of genomic RNA was electroporated into wt 251
MEFs. The cells were then treated with mouse IFN-α for 8 hours and infected with Semliki 252
Forest virus (SFV), which is highly sensitive to the antiviral activity of IFN, for 12 hours. 253
Longer RNA and incorporation of 5’GMP during the in vitro transcription reaction was used 254
to ensure proper processing of the electroporated RNA in cells by XRN1 to generate native 255
sfRNA (sfRNA1). An RNA containing IRAΔCS3 mutations (IΔRNA) and rendering it unable 256
to resist XRN1 degradation and produce full-length sfRNA1 (17) was used as a negative 257
control. Indeed, Northern blot hybridization of RNA isolated from electroporated wt MEFs 258
using an sfRNA-specific probe confirmed proper processing of electroporated RNAs into 259
sfRNA1 and sfRNA3 for sfRNA and IΔRNA, respectively (Fig. 4A). Northern blot 260
hybridization of RNA isolated from SFV-infected cells showed a decrease in the level of SFV 261
RNA accumulation in cells pre-treated with IFN compared to those not treated with IFN for 262
mock and IΔRNA electroporations; the decrease associated with IFN treatment, however, was 263
not as pronounced in cells electroporated with sfRNA (Fig. 4B). To quantify this effect, SFV 264
titers in the culture supernatant of infected cells were determined. While IFN treatment 265
reduced virus titers in mock or IΔRNA electroporated cells by 533- and 650-fold, respectively, 266
the antiviral effect of IFN was less pronounced in the presence of sfRNA and titers were only 267
reduced on average by 204-fold (Fig. 4C, P < 0.05). Thus, sfRNA conferred some resistance 268
to the antiviral activity of IFN-α independently of WNVKUN infection 269
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
13
RNase L partially restricts replication of sfRNA-deficient virus in MEFs. To begin to 270
define ISGs responsible for restricting replication of sfRNA-deficient virus we used MEFs 271
deficient in PKR or RNase L, ISGs which have documented inhibitory activity against WNV 272
in vitro and in vivo (4, 19, 25, 47, 49, 52). Replication of sfRNA-deficient virus was not 273
rescued in PKR-/- MEFs (Fig. 5A and B), but was partially rescued in RNase L-/- MEFs (Fig. 274
5C and D). However, the extend of rescue in RNase L-/- MEFs was less than that observed in 275
IFNAR-/- or IRF-3-/- x IRF-7-/- MEFs suggesting that either additional ISGs or another more 276
general mechanism induced by IFN also contributes to the restriction of replication of sfRNA-277
deficient virus. Consistent with this, the sfRNA-deficient virus remained attenuated in adult 278
RNase L-/- mice, and no mortality was observed (data not shown). 279
sfRNA fails to inhibit RNase L activity in vitro. Poliovirus (PV) competitive inhibitor 280
RNA (ciRNA) is a RNA element in PV genomic RNA (gRNA) that has been shown to inhibit 281
RNase L activity, thereby rendering PV genomic RNA resistant to cleavage by RNase L (23, 282
55). Because of the partial rescue of the sfRNA-deficient virus in RNase L-/- MEF, we 283
hypothesized that sfRNA might inhibit RNase L activity in a manner similar to that of PV 284
ciRNA and thus protect the WNV genome from RNase L mediated degradation. We initially 285
tested the sensitivity of WNV gRNA to RNase L mediated degradation in an in vitro assay 286
using recombinant RNase L and its activator 2-5A. WNV gRNA and PV RNA containing the 287
G5761A mutation that disables the ciRNA and confers sensitivity to RNase L cleavage (55, 56) 288
were almost completely degraded after 60 minutes of incubation, whereas RNase L resistant 289
PV RNA remained mostly intact (Fig. 6A, upper panel). No degradation was observed in 290
reactions without 2-5A, confirming that cleavage was RNase L specific (Fig. 6A, lower 291
panel). To assess the inhibitory potential of sfRNA on RNase L mediated cleavage of WNV 292
gRNA, we added increasing concentrations of sfRNA to the degradation assay. PV ciRNA and 293
a PV RNA lacking inhibitory capacity (PV2121) were used as positive and negative controls, 294
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
14
respectively. As expected, PV ciRNA protected WNV gRNA from RNase L mediated 295
degradation at concentrations of 500 to 2000 nM. In contrast, neither PV2121 nor sfRNA were 296
able to protect WNV RNA from degradation by RNase L even at the highest concentration of 297
2000 nM (Fig. 6B). Thus, despite the partial rescue of replication of sfRNA-deficient virus in 298
RNase L-/- MEFs, we did not observe a direct inhibitory effect by sfRNA on RNase L activity 299
in vitro. 300
301
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
15
DISCUSSION 302
We and others have shown previously that a small non-coding RNA derived from the 303
3’UTR of the flavivirus genome (sfRNA) is produced in vitro and in vivo after infection by 304
several members of the genus (29, 44, 49, 57). sfRNA is produced as a result of incomplete 305
degradation of the gRNA by a cellular ribonuclease, presumably XRN1, and is required for 306
efficient replication and virulence in mice (44). Flaviviruses are not unique in their ability to 307
generate small non-coding RNA or RNA structures within the genome that inhibit the host 308
response. Several picornaviruses including Group C enteroviruses and poliovirus encode 309
structures within their open reading frames that inhibit the endoribonuclease domain of RNase 310
L (23, 55). In addition, a number of viruses produce non-coding RNAs including micro (mi) 311
RNAs and longer non-coding (nc) RNAs (53). Viral miRNAs function in evasion of the host 312
antiviral response or in regulation of the viral life cycle (8). Longer ncRNAs also modulate 313
host antiviral responses. For example, ~160 nt adenovirus-associated RNA and ~170 nt 314
Epstein-Barr virus-encoded EBER1 and EBER2 ncRNAs inhibit PKR (6, 41). Viral ncRNAs 315
also have other functions, including maintenance of ATP levels during human 316
cytomegalovirus infection (46), or activation of T cells during Herpes virus saimiri infection 317
by inhibiting a cellular miRNA (5, 7). Given the role of viral ncRNAs in inhibiting host anti-318
viral responses and the highly attenuated phenotype of a WNVKUN mutant lacking sfRNA, we 319
hypothesized that sfRNA also might counteract the host response against flavivirus infection. 320
Type I and to a lesser extent type II IFNs control WNV and other flavivirus infections 321
(reviewed in (12)). Nonetheless, WNV has evolved several strategies that counteract IFN-322
dependent antiviral responses. WNV delays the induction of type I IFNs in infected cells by 323
uncertain mechanisms, which allows for more efficient early replication (16). In addition, 324
WNV encodes proteins with specific inhibitory functions that prevent activation of certain 325
arms of the innate immune response (60), (2, 21, 34, 35) (3, 28), (42, 43). Importantly, the 326
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
16
efficiency with which a particular WNV strain inhibits the antiviral activity of IFN is linked to 327
virulence and pathogenicity (26). A contribution of NS5 to this effect recently has been 328
established (28). In contrast to the extensive studies on the role of flavivirus proteins in 329
inhibition of the IFN response, RNA-based evasion mechanisms have only recently begun to 330
be identified (14). 331
Our results show that a mutant WNVKUN deficient in the production of sfRNA 332
replicated poorly in wt mice and in wt MEFs, whereas it replicated more efficiently in mice 333
and MEFs deficient in major factors involved in the type I IFN response. Fifty per cent 334
mortality in 17-20 day old wt C57BL/6 mice after footpad challenge with mutant virus differs 335
from our previous data that showed no mortality after intraperitoneal challenge of 19-21 day 336
old Swiss outbred mice (44). It is most likely that the difference is due to the difference in the 337
genetic background of the mice, however the different routes of inoculation and age variations 338
could also contribute to observed differences. Nevertheless, mutant virus was still significantly 339
less virulent than wt virus in wt C57BL/6 mice. In contrast, virulence of mutant virus was 340
largely rescued in mice with a combined deficiency of the transcription factors IRF-3 and IRF-341
7, which act downstream of IPS-1 and MyD88, or a deficiency of IFNAR. Both IRF-3 and 342
IRF-7 are master transcriptional regulators of the type I IFN response to WNV infection (10, 343
11). In IRF-7-/- MEFs, the IFN-α response and positive feedback loop is abolished, whereas 344
the IFN-β response remains intact (11). IRF-3-/- MEFs, in contrast, show a significant 345
reduction in early IFN-α and IFN-β responses to WNV (10) and fail to limit WNV spread (15). 346
A combined deficiency of IRF-3 and IRF-7 rescued sfRNA-deficient mutant virus almost to 347
wt virus levels, which is consistent with the finding that IRF-3-/- x IRF-7-/- MEFs had severely 348
impaired type I IFN responses after WNV infection (13). WNV is more pathogenic in IRF-3-/- 349
x IRF-7-/- mice compared to wt mice, but was still not as virulent as in IFNAR-/- mice, possibly 350
because of a small residual IFN response (13). Analogously, the virulence of the sfRNA-351
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
17
deficient virus was greater in IFNAR-/- mice than in IRF-3-/- x IRF-7-/- mice. Clearly, 352
replication of the sfRNA-deficient virus was impaired by the IFN-mediated antiviral response, 353
although remaining differences in virulence between wt and sfRNA-deficient virus in IFN 354
response-defective mice suggest also some contribution of IFN-independent antiviral response 355
in restricting virulence of sfRNA-deficient virus. 356
The greater sensitivity of sfRNA-deficient virus compared to wt virus to IFN pre-357
treatment in IRF-3-/- x IRF-7-/- MEFs and rescue of its replication in wt MEFs in the presence 358
of IFNAR-neutralizing antibodies suggests a role for sfRNA in modulating IFN-induced 359
effector function. Consistent with this, sfRNA generated in MEFs (presumably by XRN1) 360
from in vitro transcribed longer RNA with additional 5’ 288 nucleotides after electroporation 361
resulted in a partial inhibition of the IFN-induced antiviral response against an unrelated virus, 362
SFV. Interestingly, when shorter in vitro transcribed RNA representing sfRNA with only 6 363
additional nucleotides upstream of sfRNA start was used under the same conditions, no 364
inhibition of IFN response was observed (data not shown), indicating that inhibition of IFN 365
response by sfRNA may require prior or co-processing by XRN1. Although the exact 366
mechanism of sfRNA-mediated inhibition of effector function of IFN is unclear, we 367
hypothesized that sfRNA may directly interact with and antagonize an ISG that binds RNA. 368
Two of these RNA-binding ISGs, PKR and RNase L, have antiviral activity against a number 369
of viruses including WNV (4, 19, 25, 47, 49, 52). In addition to degrading single-stranded 370
RNA, which could have direct antiviral functions, RNase L can generate viral or host-derived 371
small RNAs that amplify the IFN response by generating PAMPs that activate the RIG-372
I/MDA5 pathway (37, 38). Because of these two crucial direct and indirect antiviral activities, 373
RNase L is targeted by viral evasion mechanisms including a small poliovirus RNA, termed 374
PV ciRNA (23, 55). In the current study, we observed a partial rescue of sfRNA-deficient 375
virus in RNase L-/- MEFs, suggesting that RNase L could be one of the ISG targets of sfRNA. 376
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
18
We hypothesized that sfRNA might inhibit RNase L mediated cleavage of WNV RNA in a 377
manner similar to PV ciRNA. However, our in vitro experiments showed that WNV RNA was 378
sensitive to degradation by RNase L, but that sfRNA failed to inhibit RNase L activity 379
directly. Thus, the mechanism by which sfRNA interferes with RNase L-mediated inhibition 380
of WNVKUN remains uncertain. Further detailed study on how sfRNA interferes with the 381
function of RNase L and/or other yet unidentified ISGs, as well as other IFN-dependent and 382
IFN-independent antiviral pathways is required. 383
In summary, we have demonstrated that WNVKUN sfRNA contributes to viral evasion 384
of the IFN-induced antiviral response. In addition to the many prior studies showing that viral 385
proteins can inhibit IFN induction or signaling, flaviviruses, and likely other RNA viruses, 386
independently utilize RNA-based strategies to subvert the antiviral response. An improved 387
understanding of viral evasion strategies may facilitate novel strategies for therapeutic 388
intervention against viral pathogens. Possibly, new classes of drugs that alter production of 389
sfRNA in infected cells could sensitize flaviviruses to the innate antiviral effects of type I IFN. 390
391
ACKNOWLEDGEMENTS 392
The authors would like to thank Ezequiel Balmori Melian and Judy Edmonds for 393
helpful discussions and Sarah Vick for technical assistance. This work was supported by the 394
grants from the National Health and Medical Research Council of Australia and the National 395
Institute of Health UO1 AI066321 (A.A.K.), the National Institute of Health U54 AI081680 396
(Pacific Northwest Regional Center of Excellence for Biodefense and Emerging Infectious 397
Diseases Research) and U19 AI083019 (M.S.D), and the National Institute of Health NIH 398
CA044059 (R.S). D.J.B was supported by Public Health Service grant AI042189 from the 399
NIH. H.M.L was supported by an NIH Training grant (T32-AI007172). The funders had no 400
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
19
role in study design, data collection and analysis, decision to publish, or preparation of the 401
manuscript. 402
403
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
20
REFERENCES 404
1. Andersen, J., S. VanScoy, T. F. Cheng, D. Gomez, and N. C. Reich. 2008. IRF-3-405 dependent and augmented target genes during viral infection. Genes Immun 9:168-75. 406
2. Arjona, A., M. Ledizet, K. Anthony, N. Bonafe, Y. Modis, T. Town, and E. Fikrig. 407 2007. West Nile virus envelope protein inhibits dsRNA-induced innate immune 408 responses. J Immunol 179:8403-9. 409
3. Best, S. M., K. L. Morris, J. G. Shannon, S. J. Robertson, D. N. Mitzel, G. S. Park, 410 E. Boer, J. B. Wolfinbarger, and M. E. Bloom. 2005. Inhibition of interferon-411 stimulated JAK-STAT signaling by a tick-borne flavivirus and identification of NS5 as 412 an interferon antagonist. J Virol 79:12828-39. 413
4. Brass, A. L., I. C. Huang, Y. Benita, S. P. John, M. N. Krishnan, E. M. Feeley, B. 414 J. Ryan, J. L. Weyer, L. van der Weyden, E. Fikrig, D. J. Adams, R. J. Xavier, M. 415 Farzan, and S. J. Elledge. 2009. The IFITM proteins mediate cellular resistance to 416 influenza A H1N1 virus, West Nile virus, and dengue virus. Cell 139:1243-54. 417
5. Cazalla, D., T. Yario, and J. Steitz. 2010. Down-Regulation of a Host MicroRNA by 418 a Herpesvirus saimiri Noncoding RNA. Science 328. 419
6. Clarke, P. A., M. Schwemmle, J. Schickinger, K. Hilse, and M. J. Clemens. 1991. 420 Binding of Epstein-Barr virus small RNA EBER-1 to the double-stranded RNA-421 activated protein kinase DAI. Nucleic Acids Res 19:243-8. 422
7. Cook, H. L., J. R. Lytle, H. E. Mischo, M. J. Li, J. J. Rossi, D. P. Silva, R. C. 423 Desrosiers, and J. A. Steitz. 2005. Small nuclear RNAs encoded by Herpesvirus 424 saimiri upregulate the expression of genes linked to T cell activation in virally 425 transformed T cells. Curr Biol 15:974-9. 426
8. Cullen, B. R. Viruses and microRNAs: RISCy interactions with serious consequences. 427 Genes Dev 25:1881-94. 428
9. Daffis, S., H. M. Lazear, W. J. Liu, M. Audsley, M. Engle, A. A. Khromykh, and 429 M. S. Diamond. The naturally attenuated Kunjin strain of West Nile virus shows 430 enhanced sensitivity to the host type I interferon response. J Virol 85:5664-8. 431
10. Daffis, S., M. A. Samuel, B. C. Keller, M. Gale, Jr., and M. S. Diamond. 2007. 432 Cell-specific IRF-3 responses protect against West Nile virus infection by interferon-433 dependent and -independent mechanisms. PLoS Pathog 3:e106. 434
11. Daffis, S., M. A. Samuel, M. S. Suthar, B. C. Keller, M. Gale, Jr., and M. S. 435 Diamond. 2008. Interferon regulatory factor IRF-7 induces the antiviral alpha 436 interferon response and protects against lethal West Nile virus infection. J Virol 437 82:8465-75. 438
12. Daffis, S., M. S. Suthar, M. Gale, Jr., and M. S. Diamond. 2009. Measure and 439 countermeasure: type I IFN (IFN-alpha/beta) antiviral response against West Nile 440 virus. J Innate Immun 1:435-45. 441
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
21
13. Daffis, S., M. S. Suthar, K. J. Szretter, M. Gale, Jr., and M. S. Diamond. 2009. 442 Induction of IFN-beta and the innate antiviral response in myeloid cells occurs through 443 an IPS-1-dependent signal that does not require IRF-3 and IRF-7. PLoS Pathog 444 5:e1000607. 445
14. Daffis, S., K. J. Szretter, J. Schriewer, J. Li, S. Youn, J. Errett, T. Y. Lin, S. 446 Schneller, R. Zust, H. Dong, V. Thiel, G. C. Sen, V. Fensterl, W. B. Klimstra, T. 447 C. Pierson, R. M. Buller, M. Gale Jr, P. Y. Shi, and M. S. Diamond. 2010. 2'-O 448 methylation of the viral mRNA cap evades host restriction by IFIT family members. 449 Nature 468:452-456. 450
15. Fredericksen, B. L., and M. Gale, Jr. 2006. West Nile virus evades activation of 451 interferon regulatory factor 3 through RIG-I-dependent and -independent pathways 452 without antagonizing host defense signaling. J Virol 80:2913-23. 453
16. Fredericksen, B. L., M. Smith, M. G. Katze, P. Y. Shi, and M. Gale, Jr. 2004. The 454 host response to West Nile Virus infection limits viral spread through the activation of 455 the interferon regulatory factor 3 pathway. J Virol 78:7737-47. 456
17. Funk, A., K. Truong, T. Nagasaki, S. Torres, N. Floden, E. Balmori Melian, J. 457 Edmonds, H. Dong, P. Y. Shi, and A. A. Khromykh. 2010. RNA structures required 458 for production of subgenomic flavivirus RNA. J Virol 84:11407-17. 459
18. Funk, A., K. Truong, T. Nagasaki, S. Torres, N. Floden, E. B. Melian, J. 460 Edmonds, H. Dong, P. Y. Shi, and A. A. Khromykh. 2010. RNA structures required 461 for production of subgenomic flavivirus RNA. J Virol. 462
19. Gilfoy, F. D., and P. W. Mason. 2007. West Nile virus-induced interferon production 463 is mediated by the double-stranded RNA-dependent protein kinase PKR. J Virol 464 81:11148-58. 465
20. Gubler, D. J. 2002. Epidemic dengue/dengue hemorrhagic fever as a public health, 466 social and economic problem in the 21st century. Trends Microbiol 10:100-3. 467
21. Guo, J. T., J. Hayashi, and C. Seeger. 2005. West Nile virus inhibits the signal 468 transduction pathway of alpha interferon. J Virol 79:1343-50. 469
22. Hall, R. A., A. K. Broom, D. W. Smith, and J. S. Mackenzie. 2002. The ecology and 470 epidemiology of Kunjin virus. Curr Top Microbiol Immunol 267:253-69. 471
23. Han, J. Q., H. L. Townsend, B. K. Jha, J. M. Paranjape, R. H. Silverman, and D. 472 J. Barton. 2007. A phylogenetically conserved RNA structure in the poliovirus open 473 reading frame inhibits the antiviral endoribonuclease RNase L. J Virol 81:5561-72. 474
24. Honda, K., H. Yanai, H. Negishi, M. Asagiri, M. Sato, T. Mizutani, N. Shimada, 475 Y. Ohba, A. Takaoka, N. Yoshida, and T. Taniguchi. 2005. IRF-7 is the master 476 regulator of type-I interferon-dependent immune responses. Nature 434:772-7. 477
25. Jiang, D., J. M. Weidner, M. Qing, X. B. Pan, H. Guo, C. Xu, X. Zhang, A. Birk, 478 J. Chang, P. Y. Shi, T. M. Block, and J. T. Guo. 2010. Identification of Five 479 Interferon-induced Cellular Proteins That Inhibit West Nile Virus and Dengue Virus 480 Infection. J Virol. 84:8332-41. 481
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
22
26. Keller, B. C., B. L. Fredericksen, M. A. Samuel, R. E. Mock, P. W. Mason, M. S. 482 Diamond, and M. Gale, Jr. 2006. Resistance to alpha/beta interferon is a determinant 483 of West Nile virus replication fitness and virulence. J Virol 80:9424-34. 484
27. Khromykh, A. A., A. N. Varnavski, and E. G. Westaway. 1998. Encapsidation of 485 the flavivirus kunjin replicon RNA by using a complementation system providing 486 Kunjin virus structural proteins in trans. J Virol 72:5967-77. 487
28. Laurent-Rolle, M., E. F. Boer, K. J. Lubick, J. B. Wolfinbarger, A. B. Carmody, 488 B. Rockx, W. Liu, J. Ashour, W. L. Shupert, M. R. Holbrook, A. D. Barrett, P. W. 489 Mason, M. E. Bloom, A. Garcia-Sastre, A. A. Khromykh, and S. M. Best. 2010. 490 The NS5 protein of the virulent West Nile virus NY99 strain is a potent antagonist of 491 type I interferon-mediated JAK-STAT signaling. J Virol 84:3503-15. 492
29. Lin, K. C., H. L. Chang, and R. Y. Chang. 2004. Accumulation of a 3'-terminal 493 genome fragment in Japanese encephalitis virus-infected mammalian and mosquito 494 cells. J Virol 78:5133-8. 495
30. Liu, W. J., H. B. Chen, and A. A. Khromykh. 2003. Molecular and functional 496 analyses of Kunjin virus infectious cDNA clones demonstrate the essential roles for 497 NS2A in virus assembly and for a nonconservative residue in NS3 in RNA replication. 498 J Virol 77:7804-13. 499
31. Liu, W. J., H. B. Chen, X. J. Wang, H. Huang, and A. A. Khromykh. 2004. 500 Analysis of adaptive mutations in Kunjin virus replicon RNA reveals a novel role for 501 the flavivirus nonstructural protein NS2A in inhibition of beta interferon promoter-502 driven transcription. J Virol 78:12225-35. 503
32. Liu, W. J., P. L. Sedlak, N. Kondratieva, and A. A. Khromykh. 2002. 504 Complementation analysis of the flavivirus Kunjin NS3 and NS5 proteins defines the 505 minimal regions essential for formation of a replication complex and shows a 506 requirement of NS3 in cis for virus assembly. J Virol 76:10766-75. 507
33. Liu, W. J., X. J. Wang, D. C. Clark, M. Lobigs, R. A. Hall, and A. A. Khromykh. 508 2006. A single amino acid substitution in the West Nile virus nonstructural protein 509 NS2A disables its ability to inhibit alpha/beta interferon induction and attenuates virus 510 virulence in mice. J Virol 80:2396-404. 511
34. Liu, W. J., X. J. Wang, V. V. Mokhonov, P. Y. Shi, R. Randall, and A. A. 512 Khromykh. 2005. Inhibition of interferon signaling by the New York 99 strain and 513 Kunjin subtype of West Nile virus involves blockage of STAT1 and STAT2 activation 514 by nonstructural proteins. J Virol 79:1934-42. 515
35. Mackenzie, J. M., A. A. Khromykh, and R. G. Parton. 2007. Cholesterol 516 manipulation by West Nile virus perturbs the cellular immune response. Cell Host 517 Microbe 2:229-39. 518
36. Mackenzie, J. S., D. J. Gubler, and L. R. Petersen. 2004. Emerging flaviviruses: the 519 spread and resurgence of Japanese encephalitis, West Nile and dengue viruses. Nat 520 Med 10:S98-109. 521
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
23
37. Malathi, K., B. Dong, M. Gale, Jr., and R. H. Silverman. 2007. Small self-RNA 522 generated by RNase L amplifies antiviral innate immunity. Nature 448:816-9. 523
38. Malathi, K., T. Saito, N. Crochet, D. J. Barton, M. Gale, Jr., and R. H. Silverman. 524 2010. RNase L releases a small RNA from HCV RNA that refolds into a potent 525 PAMP. RNA. 526
39. Marie, I., J. E. Durbin, and D. E. Levy. 1998. Differential viral induction of distinct 527 interferon-alpha genes by positive feedback through interferon regulatory factor-7. 528 EMBO J 17:6660-9. 529
40. Markoff, L. 2003. 5'- and 3'-noncoding regions in flavivirus RNA. Adv Virus Res 530 59:177-228. 531
41. Mathews, M. B., and T. Shenk. 1991. Adenovirus virus-associated RNA and 532 translation control. J Virol 65:5657-62. 533
42. Mazzon, M., M. Jones, A. Davidson, B. Chain, and M. Jacobs. 2009. Dengue virus 534 NS5 inhibits interferon-alpha signaling by blocking signal transducer and activator of 535 transcription 2 phosphorylation. J Infect Dis 200:1261-70. 536
43. Munoz-Jordan, J. L., M. Laurent-Rolle, J. Ashour, L. Martinez-Sobrido, M. 537 Ashok, W. I. Lipkin, and A. Garcia-Sastre. 2005. Inhibition of alpha/beta interferon 538 signaling by the NS4B protein of flaviviruses. J Virol 79:8004-13. 539
44. Pijlman, G. P., A. Funk, N. Kondratieva, J. Leung, S. Torres, L. van der Aa, W. J. 540 Liu, A. C. Palmenberg, P. Y. Shi, R. A. Hall, and A. A. Khromykh. 2008. A highly 541 structured, nuclease-resistant, noncoding RNA produced by flaviviruses is required for 542 pathogenicity. Cell Host Microbe 4:579-91. 543
45. R-Development-Core-Team. 2010. R: A language and environment for 544 statistical computing. R Foundation for Statistical Computing, Vienna, Austria. 545
46. Reeves, M. B., A. A. Davies, B. P. McSharry, G. W. Wilkinson, and J. H. Sinclair. 546 2007. Complex I binding by a virally encoded RNA regulates mitochondria-induced 547 cell death. Science 316:1345-8. 548
47. Samuel, M. A., K. Whitby, B. C. Keller, A. Marri, W. Barchet, B. R. Williams, R. 549 H. Silverman, M. Gale, Jr., and M. S. Diamond. 2006. PKR and RNase L contribute 550 to protection against lethal West Nile Virus infection by controlling early viral spread 551 in the periphery and replication in neurons. J Virol 80:7009-19. 552
48. Sato, M., H. Suemori, N. Hata, M. Asagiri, K. Ogasawara, K. Nakao, T. Nakaya, 553 M. Katsuki, S. Noguchi, N. Tanaka, and T. Taniguchi. 2000. Distinct and essential 554 roles of transcription factors IRF-3 and IRF-7 in response to viruses for IFN-alpha/beta 555 gene induction. Immunity 13:539-48. 556
49. Scherbik, S. V., J. M. Paranjape, B. M. Stockman, R. H. Silverman, and M. A. 557 Brinton. 2006. RNase L plays a role in the antiviral response to West Nile virus. J 558 Virol 80:2987-99. 559
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
24
50. Sheehan, K. C., K. S. Lai, G. P. Dunn, A. T. Bruce, M. S. Diamond, J. D. Heutel, 560 C. Dungo-Arthur, J. A. Carrero, J. M. White, P. J. Hertzog, and R. D. Schreiber. 561 2006. Blocking monoclonal antibodies specific for mouse IFN-alpha/beta receptor 562 subunit 1 (IFNAR-1) from mice immunized by in vivo hydrodynamic transfection. J 563 Interferon Cytokine Res 26:804-19. 564
51. Silva, P. A., C. F. Pereira, T. J. Dalebout, W. J. Spaan, and P. J. Bredenbeek. 565 2010. An RNA pseudoknot is required for production of yellow fever virus 566 subgenomic RNA by the host nuclease XRN1. J Virol 84:11395-406. 567
52. Silverman, R. H. 2007. Viral encounters with 2',5'-oligoadenylate synthetase and 568 RNase L during the interferon antiviral response. J Virol 81:12720-9. 569
53. Sullivan, C. S. 2008. New roles for large and small viral RNAs in evading host 570 defences. Nat Rev Genet 9:503-7. 571
54. Takeuchi, O., and S. Akira. 2009. Innate immunity to virus infection. Immunol Rev 572 227:75-86. 573
55. Townsend, H. L., B. K. Jha, J. Q. Han, N. K. Maluf, R. H. Silverman, and D. J. 574 Barton. 2008. A viral RNA competitively inhibits the antiviral endoribonuclease 575 domain of RNase L. Rna 14:1026-36. 576
56. Townsend, H. L., B. K. Jha, R. H. Silverman, and D. J. Barton. 2008. A putative 577 loop E motif and an H-H kissing loop interaction are conserved and functional features 578 in a group C enterovirus RNA that inhibits ribonuclease L. RNA Biol 5:263-72. 579
57. Urosevic, N., M. van Maanen, J. P. Mansfield, J. S. Mackenzie, and G. R. 580 Shellam. 1997. Molecular characterization of virus-specific RNA produced in the 581 brains of flavivirus-susceptible and -resistant mice after challenge with Murray Valley 582 encephalitis virus. J Gen Virol 78 ( Pt 1):23-9. 583
58. Westaway, E. G., J. M. Mackenzie, and A. A. Khromykh. 2003. Kunjin RNA 584 replication and applications of Kunjin replicons. Adv Virus Res 59:99-140. 585
59. Westaway, E. G., J. M. Mackenzie, and A. A. Khromykh. 2002. Replication and 586 gene function in Kunjin virus. Curr Top Microbiol Immunol 267:323-51. 587
60. Wilson, J. R., P. F. de Sessions, M. A. Leon, and F. Scholle. 2008. West Nile virus 588 nonstructural protein 1 inhibits TLR3 signal transduction. J Virol 82:8262-71. 589
61. Zhou, A., J. Paranjape, T. L. Brown, H. Nie, S. Naik, B. Dong, A. Chang, B. 590 Trapp, R. Fairchild, C. Colmenares, and R. H. Silverman. 1997. Interferon action 591 and apoptosis are defective in mice devoid of 2',5'-oligoadenylate-dependent RNase L. 592 EMBO J 16:6355-63. 593
594 595
596
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
25
FIGURE LEGENDS 597
Figure 1. Replication of sfRNA-deficient WNVKUN is rescued in IRF-3-/- x IRF-7-/- MEFs. 598
(A) Northern blot of viral RNA using a 3’UTR-specific probe in wt and IRF-3-/- x IRF-7-/- 599
MEFs infected with MOI of 1 of wt (FLSDX) and sfRNA-deficient (FL-IRAΔCS3) viruses. 600
Ethidium bromide staining of ribosomal RNA is included as loading control. (B) 601
Corresponding titers of infectious virus in the supernatant of the infected cells. The results are 602
the average of 2 (wt MEFs) or 4 (IRF-3-/- x IRF-7-/- MEFs) independent experiments. Error 603
bars indicate standard deviation. (C) Virus titers in the supernatants of IRF-3-/- x IRF-7-/- 604
MEFs infected with MOI of 0.1 and 0.01 of wt (FLSDX) or sfRNA-deficient (FL-IRAΔCS3) 605
viruses. gRNA, genomic RNA; sfRNA, subgenomic flavivirus RNA; PFU, plaque-forming 606
units; hpi, hours post infection. 607
Figure 2. sfRNA-deficient virus is more sensitive to IFNα and its replication is rescued by 608
neutralizing antibodies to IFNAR. (A) Northern blot of viral RNA in IRF-3-/- x IRF-7-/- 609
MEFs infected with wt (FLSDX) or sfRNA-deficient (FL-IRAΔCS3) viruses after pre-610
treatment with IFN-α (0 to 1000 IU/ml for 8 hours prior to infection at MOI of 1). 611
Supernatants and cells were harvested 48 hpi for plaque assay and Northern blot with a 3’UTR 612
specific probe. Ethidium bromide staining of ribosomal RNA is included as loading control. 613
(B) Corresponding titers of infectious virus in the supernatant of the infected cells. Results are 614
representative of two independent experiments. (C) Northern blot of viral RNA in wt MEFs 615
infected with wt (FLSDX) and sfRNA-deficient (FL-IRAΔCS3) viruses in the presence of 616
indicated concentrations of IFNAR-neutralizing antibody MAR1-5A3 or isotype control 617
antibody. Wt MEFs were infected with the wt and mutant viruses for 2 hours at MOI of 1 after 618
which MAR1-5A3 or an isotype control antibody was added. Cells were harvested four days 619
later, and Northern blot was performed to detect viral RNA using a 3’UTR specific probe. 620
Ethidium bromide staining of ribosomal RNA is included as loading control. 621
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
26
Figure 3. Rescue of pathogenesis of sfRNA-deficient virus in mice deficient in IFN 622
induction or signaling. A. Weanling (17 to 20 day-old) wt C57BL/6 mice were inoculated 623
with 103 PFU of wt (FLSDX) or sfRNA-deficient (FL-IRAΔCS3) viruses by footpad injection 624
and followed for mortality for 21 days. Survival data were combined from three independent 625
experiments with a total of 16 to 18 mice per group (P < 0.002). B. Adult (8 to 10 week-old) 626
IRF-3-/- x IRF-7-/- mice were inoculated with 103 PFU of wt (FLSDX) or sfRNA-deficient 627
(FL-IRAΔCS3) viruses by footpad injection and followed for mortality for 21 days. Survival 628
data was combined from 2 independent experiments with a total of 6 to 10 mice per group (P < 629
0.05). C - E. Adult (8 to 10 week-old) IFNAR-/- mice were inoculated with 103 (C), 102 (D), 630
or 101 (E) PFU of wt (FLSDX) or sfRNA-deficient (FL-IRAΔCS3) viruses by footpad 631
injection and followed for mortality. Survival data was combined from 3 independent 632
experiments with a total of 5 to 9 mice per group (103 PFU: P < 0.02; 102 and 101 PFU: P < 633
0.001). F - I. Viral RNA levels were determined from serum samples harvested on the 634
indicated days after infection of weanling mice with 103 PFU, or adult IFNAR-/- mice with 103, 635
102 or 101 PFU of wt (FLSDX) or sfRNA-deficient (FL-IRAΔCS3) viruses using qRT-PCR. 636
Data are shown as log10 viral RNA equivalents per ml from 11 to 18 (wt) or 4 to 9 (IFNAR-/-) 637
mice per time point. The error bars indicate standard error of the mean, and the dotted line 638
represents the limit of sensitivity of the assay. Asterisks and corresponding p values shown 639
represent differences that are statistically significant by 2-way ANOVA. 640
Figure 4. sfRNA reduces the inhibitory effect of IFN treatment on SFV replication. Wild 641
type MEFs were electroporated with in vitro transcribed sfRNA or sfRNA containing 642
IRAΔCS3 mutations (IΔRNA) and treated with 1,000 IU/ml mouse IFN-α for 8 hours. Cells 643
were infected subsequently with SFV at an MOI of 1 for one hour. Samples were harvested 644
10h post electroporation (before SFV infection) and 12h after SFV infection. (A) Northern 645
blot of RNA isolated before SFV infection showing different sfRNA species using a 3’UTR-646
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
27
specific probe. RNA from cells infected with FLSDX (F) and FL-IRAΔCS3 (IΔ) viruses were 647
used as controls for detecting sfRNA 1 and sfRNA3, respectively. Ethidium bromide staining 648
of ribosomal RNA (rRNA) is included as loading control. (B) Northern blot detecting SFV 649
gRNA with a probe for SFV-nsp4 gene. Ethidium bromide staining of ribosomal RNA is 650
included as loading control. Representative blot from three independent experiments is shown. 651
(C) Fold reduction of SFV titers after IFN treatment in the presence of sfRNA or IΔRNA. 652
Mean of three independent experiments is shown. Error bars indicate standard error of the 653
mean. Statistical significance as analysed by unpaired t test is indicated by asterisks (*, P < 654
0.05). 655
Figure 5. Replication of sfRNA mutant WNVKUN is partially rescued in RNase L-/- but not 656
in PKR-/- MEFs. (A) Northern blot of viral RNA in wt (top) and PKR-/- (bottom) MEFs 657
infected with wt (FLSDX) and sfRNA-deficient (FL-IRAΔCS3) viruses. (B) Corresponding 658
titres of infectious virus in the supernatant of the infected cells. The data is the average of 2 659
independent experiments. Error bars indicate standard deviation. (C) Northern blot of viral 660
RNA in wt MEFs (top) or RNase L-/- MEFs (bottom) infected with wt (FLSDX) and sfRNA-661
deficient (FL-IRAΔCS3) viruses. (D) Corresponding titres of infectious virus in the 662
supernatant of the infected cells. The data is the average of 2 independent experiments 663
performed in duplicate. Error bars indicate standard deviation. Viral RNA was detected in all 664
northern blots with a 3’UTR-specific probe. Ethidium bromide staining of ribosomal RNA is 665
included as loading control for each blot. 666
Figure 6. WNV RNA is sensitive to RNase L degradation in vitro and sfRNA fails to 667
protect WNV RNA from RNase L degradation. (A) 150 nM of radiolabeled RNase L-668
resistant poliovirus (PV) RNA, RNase L-sensitive PV G5761A RNA and WNV RNA were 669
incubated for the indicated periods of time in reactions containing 20 nM RNase L and 20 nM 670
2-5A (upper panel) or 20 nM RNase L without 2-5A (lower panel). Reactions were terminated 671
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from
28
in SDS buffer, phenol-chloroform extracted, ethanol precipitated with tRNA carrier, and 672
fractionated by electrophoresis in 1.2% agarose MOPS formaldehyde. RNA was detected by 673
ethidium bromide staining and UV light (left panels) and by phosphorimaging (right panels). 674
(B) Radiolabeled WNV RNA (50nM) was incubated for 60 minutes in reactions containing 20 675
nM RNase L and 20 nM 2-5A (lanes 2-11) or 20 nM RNase L without 2-5A (lane 1). 500, 676
1000, and 2000 nM sfRNA (lanes 3-5), PV2122 ciRNA (lanes 6-8) and PV2121 RNA (lanes 9-677
11) were included in the reactions. Reactions were terminated in SDS buffer, phenol-678
chloroform extracted, ethanol precipitated with tRNA carrier, and fractionated by 679
electrophoresis in 1.2% agarose MOPS formaldehyde. RNA was detected by ethidium 680
bromide staining and UV light (left panel) and by phosphorimaging (right panel). The 681
mobility of sfRNA and poliovirus RNAs is indicated with asterisks in the left panel. 682
on June 30, 2018 by guesthttp://jvi.asm
.org/D
ownloaded from