Activation of 2′ 5′-Oligoadenylate Synthetase by Stem Loops at the 5′-End of the West Nile Virus Genome

Abstract

West Nile virus (WNV) has a positive sense RNA genome with conserved structural elements in the 5' and 3' -untranslated regions required for polyprotein production. Antiviral immunity to WNV is partially mediated through the production of a cluster of proteins known as the interferon stimulated genes (ISGs). The 2' 5'-oligoadenylate synthetases (OAS) are key ISGs that help to amplify the innate immune response. Upon interaction with viral double stranded RNA, OAS enzymes become activated and enable the host cell to restrict viral propagation. Studies have linked mutations in the OAS1 gene to increased susceptibility to WNV infection, highlighting the importance of OAS1 enzyme. Here we report that the region at the 5'-end of the WNV genome comprising both the 5'-UTR and initial coding region is capable of OAS1 activation in vitro. This region contains three RNA stem loops (SLI, SLII, and SLIII), whose relative contribution to OAS1 binding affinity and activation were investigated using electrophoretic mobility shift assays and enzyme kinetics experiments. Stem loop I, comprising nucleotides 1-73, is dispensable for maximum OAS1 activation, as a construct containing only SLII and SLIII was capable of enzymatic activation. Mutations to the RNA binding site of OAS1 confirmed the specificity of the interaction. The purity, monodispersity and homogeneity of the 5'-end (SLI/II/III) and OAS1 were evaluated using dynamic light scattering and analytical ultra-centrifugation. Solution conformations of both the 5'-end RNA of WNV and OAS1 were then elucidated using small-angle x-ray scattering. In the context of purified components in vitro, these data demonstrate the recognition of conserved secondary structural elements of the WNV genome by a member of the interferon-mediated innate immune response.

Full text

Activation of 2959-Oligoadenylate Synthetase by Stem
Loops at the 59-End of the West Nile Virus Genome
Soumya Deo
1
, Trushar R. Patel
1
, Edis Dzananovic
1
, Evan P. Booy
1
, Khalid Zeid
1
, Kevin McEleney
1,4
,
Stephen E. Harding
2
, Sean A. McKenna
1,3
*
1Department of Chemistry, University of Manitoba, Winnipeg, Manitoba, Canada, 2National Centre for Macromolecular Hydrodynamics, University of Nottingham, Sutton
Bonington, United Kingdom, 3Department of Biochemistry and Medical Genetics, University of Manitoba, Winnipeg, Manitoba, Canada, 4Manitoba Institute for Materials,
University of Manitoba, Winnipeg, Manitoba, Canada
Abstract
West Nile virus (WNV) has a positive sense RNA genome with conserved structural elements in the 59and 39-untranslated
regions required for polyprotein production. Antiviral immunity to WNV is partially mediated through the production of a
cluster of proteins known as the interferon stimulated genes (ISGs). The 2959-oligoadenylate synthetases (OAS) are key ISGs
that help to amplify the innate immune response. Upon interaction with viral double stranded RNA, OAS enzymes become
activated and enable the host cell to restrict viral propagation. Studies have linked mutations in the OAS1 gene to increased
susceptibility to WNV infection, highlighting the importance of OAS1 enzyme. Here we report that the region at the 59-end
of the WNV genome comprising both the 59-UTR and initial coding region is capable of OAS1 activation in vitro. This region
contains three RNA stem loops (SLI, SLII, and SLIII), whose relative contribution to OAS1 binding affinity and activation were
investigated using electrophoretic mobility shift assays and enzyme kinetics experiments. Stem loop I, comprising
nucleotides 1-73, is dispensable for maximum OAS1 activation, as a construct containing only SLII and SLIII was capable of
enzymatic activation. Mutations to the RNA binding site of OAS1 confirmed the specificity of the interaction. The purity,
monodispersity and homogeneity of the 59-end (SLI/II/III) and OAS1 were evaluated using dynamic light scattering and
analytical ultra-centrifugation. Solution conformations of both the 59-end RNA of WNV and OAS1 were then elucidated
using small-angle x-ray scattering. In the context of purified components in vitro, these data demonstrate the recognition of
conserved secondary structural elements of the WNV genome by a member of the interferon-mediated innate immune
response.
Citation: Deo S, Patel TR, Dzananovic E, Booy EP, Zeid K, et al. (2014) Activation of 2959-Oligoadenylate Synthetase by Stem Loops at the 59-End of the West Nile
Virus Genome. PLoS ONE 9(3): e92545. doi:10.1371/journal.pone.0092545
Editor: Eric Jan, University of British Columbia, Canada
Received April 11, 2013; Accepted February 25, 2014; Published March 20, 2014
Copyright: ß2014 Deo et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted
use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This work was supported by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada (NSERC) and the Manitoba
Health Research Council (MHRC). ED was partially supported by the University of Manitoba GETS program. EPB was support by a Manitoba Health Research
Council Postdoctoral Fellowship. SD was supported by a University of Manitoba Graduate Fellowship and Faculty of Science Scholarships. KZ was supported by
the NSERC USRA program. TRP was supported by a Canadian Institutes of Health Research Postdoctoral Fellowship. The funders had no role in study design, data
collection and analysis, decision to publish, or preparation of the manuscript.
Competing Interests: The authors have declared that no competing interests exist.
* E-mail: Sean.mckenna@umanitoba.ca
Introduction
West Nile virus (WNV) is an 11 kb positive-sense, single-
stranded RNA virus that belongs to the Flaviviridae family. This
family includes known pathogenic viruses that cause Yellow fever,
Dengue fever, Japanese encephalitis, and Tick-borne encephalitis
[1–5]. The viral genome consists of a single open reading frame
(ORF) that encodes a large polyprotein precursor containing both
structural and non-structural proteins [6–9]. The ORF is flanked
upstream and downstream by the 59and 39-untranslated regions
(UTRs) which, based on thermodynamic predictions, are rich in
stable secondary structures and highly conserved amongst
Flaviviridae family members despite the lack of extensive sequence
homology [9–12]. Regions in both the 59and 39-UTRs are
necessary for translation initiation and minus strand RNA
synthesis [13,14]. The 59-UTR and downstream initial coding
region (SLI/II/III; nucleotides 1–146) is comprised of three stem-
loops (SLI; nucleotides 1–73, SLII; nucleotides 73–110, and SLIII
nucleotides 111–146), of which SLII contains the AUG start codon
for polyprotein translation (Fig. 1). RNase probing experiments
confirmed these predicted secondary structures [11]. Furthermore,
WNV non-structural protein 5, a methyltransferase, binds
specifically to the SLI of genomic RNA, and this structure is
essential for viral RNA genome replication [11]. RNase probing of
the dengue virus 59-UTR regions demonstrates similar secondary
structure interactions in SLA (equivalent to SLI in WNV),
suggesting a common structural arrangement of the region
amongst flavivirus family members [15,16].
Two long distance interactions mediated by complementary
base pairing between the 59-UTR/initial coding region and 39-
UTR are thought to be required for WNV genome cyclization,
creating a panhandle structure. Both the 59-UAR (upstream
initiation AUG region, in SLII) and 59-CS (conserved sequence, in
SLIII base pair with their complimentary sequence from regions in
the 39-UTR (39-UAR and 39-CS respectively) to achieve
cyclization. Panhandle formation enables recruitment of factors,
including the viral RNA dependent RNA polymerase [13,14,17],
that are required for minus-strand RNA synthesis. Taken together,
PLOS ONE | www.plosone.org 1 March 2014 | Volume 9 | Issue 3 | e92545

the structural features of the WNV SLI/II/III and other
Flaviviridae family members represents a highly structured and
important regulatory region.
The innate immune system is the first line of defense against
viral infection, and susceptibility of IFN a/breceptor deficient
mice to WNV infection points towards the importance of antiviral
immunity conferred by the Type 1 IFN [18,19]. Virulence and
pathogenicity of certain WNV strains has also been strongly linked
to resistance to IFN [20]. For the interferon response activation,
cellular pattern recognition receptors (PRRs) recognize pathogen
associated molecular patterns including viral genomic RNA
secondary structures. Protein PRRs involved in viral dsRNA
recognition include the retinoic acid inducible gene-1 (RIG-I),
melanoma differentiation associated gene 5 (MDA-5), double
stranded RNA-activated protein kinase (PKR), 2959-oligoadeny-
late synthetases (OAS) and adenosine deaminase acting on RNAs
(ADARs) [21,22]. The interferon response mediates its antiviral
effect by up-regulating the transcription of interferon-stimulated
genes (ISGs) leading to a significant increase in production of
various antiviral effector proteins. The 2959-oligoadenylate
synthetases (OAS) are key ISG effector proteins that help to
amplify innate immune response to viral infection [23]. The
common mode of action of family members of nucleotidyl
transferases is that, upon interaction with dsRNA, become
activated to polymerize ATP into unusual oligoadenylate chains
[29-59(A)] where the 29carbon of ribose sugar of an adenosine
mono-phosphate is linked to the 59carbon of the next [24,25].
dsRNA binding results in a conformational change that properly
orients an aspartic acid triad necessary for catalysis in the active
site of OAS enzymes [26]. These oligoadenylate chains bind to
and activate the endoribonuclease RNAse L, which destroys all
single-stranded RNA including viral RNA, thereby attenuating
viral protein production [27,28].
Several lines of evidence suggest that OAS enzymes play a key
role in the IFN response to WNV infection. The importance of
OAS enzymes in the antiviral response against WNV can be
inferred from the finding that RNase L limits WNV spread in
mouse models [29]. Sangster et al. [30] demonstrated that OAS1
and other OAS isoforms are able to reduce flavivirus yields by
99%. Additionally, a single nucleotide polymorphism (SNP) in the
OAS1 gene acts as a host genetic risk factor for humans in WNV
infection [31]. A second SNP in humans also established the
OAS1 gene as a potential genetic risk factor in WNV infection and
progression of the disease [32]. The antiviral effect of OAS
proteins have also been demonstrated against picornavirus, which
has a positive single stranded RNA genome similar to WNV [33].
Human OAS1 isotypes p42 and p46 have been shown in human
cell lines to block, in a RNase L dependent manner, the viral
replication of Dengue virus, which also belongs to family
Flaviviridae [34]. Taken together, these results suggest that OAS1
may play a role in human antiviral response to WNV infection.
To date, no specific OAS1 recognition sites have been identified
within the WNV RNA genome. Given the importance of OAS
enzymes to restrict viral propagation via dsRNA binding, we
sought to identify and analyze specific dsRNA regions of the WNV
genome responsible for OAS1 activation. dsRNAs with stable
secondary structure are ideal activators of OAS1. Regions
including the 59-UTR/initial coding regions and the 39-UTR
from members of the Flaviviridae family form conserved dsRNA
stem loops that are necessary for the regulation of viral genome
replication [9,11]. We initiated studies do determine whether a
specific RNA construct at the 59-end of the WNV genome (SLI/
II/III) could activate OAS1 enzymatic activity. In the current
study, we demonstrate the in vitro activation of OAS1 by the SLI/
II/III of WNV. Small-angle X-ray scattering experiments
demonstrated that the WNV SLI/II/III adopts multiple confor-
mations, including a subset that supports the predicted secondary
structure. Truncations to the SLI/II/III were used to narrow the
minimal region required for OAS1 activation, and mutations in
the RNA binding site of OAS1 confirmed the specificity of the
interaction [26,35]. Taken together, the results presented suggest
the biophysical basis for the regulation of OAS1 enzymatic activity
by conserved secondary structural elements at the 59-end of the
WNV RNA genome.
Figure 1. Secondary structure of the WNV 59-end. Highlighted are SLI, SLII, SLIII, the AUG start codon (black circles), the upstream AUG region
(SLI/II/III, solid line) and the conserved sequence element (59-CS, solid line) [11].
doi:10.1371/journal.pone.0092545.g001
Regulation of OAS1 by the 59-UTR of the WNV Genome
PLOS ONE | www.plosone.org 2 March 2014 | Volume 9 | Issue 3 | e92545

Materials and Methods
Expression and purification of recombinant human OAS1
Expression of recombinant human OAS1 p42 isoform (tran-
script variant 2) as a fusion protein in BL21 DE3-RIL cells
(Invitrogen, USA) and subsequent purification of the free protein
from the cleaved N-terminal affinity tag (GNSHT: GB1, Nus A,
Streptavidin, 6xHis and TEV protease site) was performed as
described previously [36]. Point mutants of OAS1 (S162G,
R195E, K199E) were generated using site-specific primers
designed using the Quikchange primer design program (Agilent
technologies, USA). The Quikchange kit (Agilent technologies,
USA) was used to generate overexpression plasmids with the
desired mutation. Plasmids were confirmed by sequencing and
then expressed/purified in the same way as described for wild type
OAS1 [36].
In vitro transcription and purification of RNA
The SLI/II/III encompassing the 59-UTR and initial coding
region of WNV (NY99iso-1, nucleotides 1–146) was generated by
in vitro transcription from a linearized plasmid using a non-
denaturing column chromatography approach [37,38]. RNA
homogeneity was assessed by denaturing Tris borate-EDTA
polyacrylamide gel electrophoresis with 8 M urea after mixing
with an equal volume of 2 X denaturing loading buffer (95%
Formamide, 18 mM EDTA, 0.01% Xylene Cyanol, 0.02%
Bromophenol Blue) and heated at 95 uC for 5 minutes in 1 X
TBE buffer (89 mM Tris Base, 89 mM Boric acid, 2 mM EDTA,
pH 8.0). The RNA concentration was determined by spectropho-
tometry at 260 nm. The following truncations of the SLI/II/III
region were also produced in an identical manner: SLI (1–73 with
an additional 59G), SLII (73–110 with an additional 59G), SLIII
(111–146), SLI/II (1–109 with an additional 59G), and SLII/III
(73–146 with an additional 59G). A ssRNA (TCTCAAAGAAA-
CACGTGCCGCTTACGCCCACAGTGTTCT) was tran-
scribed/purified in an identical manner to serve as a negative
control and to ensure that no unintended by-products of the
transcription reaction were leading to activation.
Analytical ultracentrifugation (AUC)
The sedimentation velocity (SV) experiment for OAS1 was
performed using an Optima XL-I analytical ultracentrifuge with
an An60-Ti rotor at 20.0 uC as described previously [39].
Standard 12 mm double sector cells were used where OAS1 [0.4,
0.6 and 0.8 mg/mL in 50 mM Tris, 100 mM NaCl and 1 mm
DTT (pH 7.5)] and buffer were loaded in appropriate channels.
SV data were collected at 7-minute intervals at 280 nm and
45,000 rpm using an absorption optical system. Data were
analyzed using the SEDFIT program [40,41] to obtain the
sedimentation coefficients at each concentration (s
20,b
) which were
then corrected to standard solvent conditions (s
20,w
) using the
algorithm SEDNTERP [42]. The s
20,w
(S) values for individual
concentrations were then extrapolated to infinite dilution to obtain
s
0
20,w
(S).
Dynamic light scattering (DLS)
Dynamic light scattering data were collected using a Zetasizer
Nano S system (Malvern instruments Ltd, Malvern, UK) as
described previously [43]. A scattering angle of 173uwas
employed. Wild type OAS1 in 50 mM Tris (pH 7.5), 100 mM
NaCl and 1 mM DTT was filtered using a 0.1 mm syringe filter
(Millipore, USA) and subjected to DLS measurements at 20.0 uC
at 3 different concentrations. Similarly, the DLS data for the
WNV SLI/II/III was collected at a single concentration in
50 mM Tris (pH 7.0), 100 mM NaCl. The molecular weight of
OAS1 was calculated using a version of the Svedberg equation
adjusted to include the equivalent hydrodynamic radius r
H
in place
of the translational diffusion coefficient:
Mw~
6pg0rHNs0
20,w
1{v
{
r0
ð1Þ
where 
nn is the partial specific volume, g
o
is the solvent viscosity,
r
o
is the solvent density and Nis the Avogadro’s number.
Small angle X-ray scattering (SAXS)
SAXS data for proteins (wild type and mutants) and for the
WNV SLI/II/III was collected using an in-house Rigaku
instrument as described previously [44]. SAXS data for wild type
OAS1, R195E and K199E mutants were collected at multiple
concentrations (wild type: 3.1, 3.8, 4.5 and 5.2 mg/mL; R195E:
2.2, 2.6, 3.0 and 3.4 mg/mL; K199E: 2.3, 2.7, 3.1 and 3.5 mg/
mL) in 50 mM Tris, 100 mM NaCl and 1 mM DTT at pH 7.5.
SAXS data for the SLI/II/III (in 50 mM Tris, 100 mM NaCl and
20 mM MgCl
2
at pH 7.0) were also collected at multiple
concentrations (0.8 mg/mL, 1.6 mg/mL and 2.0 mg/mL). Pri-
mary data analysis was performed using the program PRIMUS
[45], followed by estimation of the root mean square radius of
gyration (r
G
) and the maximum particle dimension (D
max
) using the
program GNOM [46]. Ab initio shape reconstruction of OAS1 WT
was performed using the program DAMMIF, that utilizes
simulated annealing protocol [47]. In addition to the ab initio
shape determination, high-resolution structure of human OAS1
(PDB code: 4IG8 [26]) was used to reconstruct the solution
conformation of OAS1 using the program BUNCH [48] as
described earlier [49]. Twelve models using DAMMIF and ten
models using BUNCH were generated which were then rotated,
aligned and averaged using DAMAVER [50]. The program
HYDROPRO [51] was employed to calculate solution properties
such as hydrodynamic radius, radius of gyration and maximal
particle dimension for each model calculated using SAXS data
following a similar approach as outlined previously [44]. The input
parameters included the density (1.0038 g/mL) and viscosity
(0.01026 Poise) of buffer as well as partial specific volume of OAS1
(0.7424 mL/g), obtained from the program SEDNTERP [42].
The molecular weight of OAS1 was calculated from its amino-acid
sequence using the protparam utility on Expasy server [52].
Electrophoretic mobility shift assay (EMSA)
RNA (100 nM) was titrated with increasing concentration of
OAS1, in 50 mM Tris, 100 mM NaCl (pH 7.5) buffer for EMSA
experiments. The binding reaction was allowed to proceed for 10
minutes at room temperature (,20.0uC) and the mixed with
native loading buffer (to a final concentration of 0.02%
bromophenol blue, 0.01% xylene cyanol FF and 1% glycerol in
1X TBE buffer) was added. Protein-RNA complex formation was
analyzed on a Tris borate-EDTA poly-acrylamide gel (8%), and
electrophoresis was performed at 65 V at ,4.0uC in 0.5X TBE
running buffer. Sybr gold (Invitrogen, USA) was used to visualize
the RNA-containing species on the gel.
OAS1 activity assay
OAS1 activity was measured using an established colorimetric
assay that quantifies the amount of pyrophosphate (PPi) produced
as a by-product of 2959-oligoadenylates formation by the active
enzyme [36]. Briefly, reaction velocities (V) were calculated by
Regulation of OAS1 by the 59-UTR of the WNV Genome
PLOS ONE | www.plosone.org 3 March 2014 | Volume 9 | Issue 3 | e92545

linear regression within the linear range of time course in at least
triplicate. The apparent dissociation constant (K
app
) and the
maximum reaction velocity (V
max
) were determined using the
following equation: V=V
max
/(1+(K
app
/[RNA])) [53]. OAS1 con-
centrations of 300 nM (for comparative activation assays) and
400 nM (for kinetic studies) were used as they fall within the
concentration range from 50 to 400 nM previously been
determined as optimal for kinetic analysis [36].
Results
Solution conformation of recombinant human OAS1
In order to study the regulation of OAS1 activity by the SLI/II/
III of the WNV genome, we first characterized the solution
properties of the human recombinant protein to ensure homoge-
neity. Sedimentation velocity experiments using an analytical
ultracentrifuge on purified OAS1 WT produced a single peak with
a sedimentation coefficient value of 3.2660.05 S (Svedberg units,
S=10
213
sec) suggesting that the protein is homogenous in mass
and conformation (Fig. 2A). The homogeneity of OAS1 was
further studied using DLS at multiple concentrations that provided
the hydrodynamic radius (r
H
) of 3.060.3 nm for OAS1 (Fig. 2B).
By taking the advantage of AUC and DLS data, an average
molecular weight of 43.0 kDa was calculated for OAS1 that agrees
well with the calculated molecular weight of 41.2 kDa. The results
support the observation that OAS1 (p42 isotype) synthesized by
cell free translation has been previous reported as monomeric [54].
A summary of all hydrodynamic properties for OAS1 is presented
in Table 1.
Next, the solution conformation of recombinant human OAS1
was determined. SAXS data were collected at multiple concen-
trations and merged to obtain a single output file (inset – Fig. 2C).
A maximum particle dimension (D
max
) of 7.1 nm and a radius of
gyration (r
G
) of 2.2860.02 nm were obtained for OAS1 from the
pair distribution function analysis (Fig. 2C). The ab initio shape
reconstruction of OAS1 was performed and the goodness of fit
parameter (xvalue) of ,0.9 was obtained for each individual
model, signifying excellent agreement between the experimental
scattering data and the calculated scattering data. The superim-
posed ab initio models provided an averaged model that was highly
similar to each individual model in terms of shape as evidenced by
normalized spatial discrepancy (NSD) parameter of 0.5260.02
(Fig. 2D). The recently determined high-resolution structure of
human OAS1 superimposed almost perfectly on the averaged ab
initio model of OAS1 [26] (Fig. 2D). We additionally validated our
ab initio modeling approach using the program BUNCH, that
generated solution conformations of OAS1 based on existing high-
resolution structural information that compared favorably with the
ab initio models (Table 1). Furthermore, excellent agreement was
observed between the experimentally determined hydrodynamic
parameters from AUC, DLS and SAXS and parameters
calculated from ab initio and BUNCH models of solution
conformations (Table 1) using the program HYDROPRO.
Hydrodynamic parameters calculated based on the previously
published high-resolution structure of OAS1 are also in good
agreement with the SAXS-derived models.
Solution conformation of the SLI/II/III of WNV
While RNAse probing experiments are consistent with the
predicted secondary structure for the SLI/II/III of WNV, the
three-dimensional structure of this RNA region is not known. We
therefore in vitro transcribed WNV SLI/II/III (nucleotides 1–146,
including the 59-UTR and initial coding region) for the purpose of
determining the solution structure by SAXS. Denaturing gel
electrophoresis demonstrated a single band of appropriate size
(data not shown), and DLS analysis confirmed that the sample was
monodisperse with a r
H
of 5.160.2 nm (Fig. 3A). Raw SAXS data
acquired at multiple concentrations were merged (inset Fig. 3B)
and the pair distribution function analysis yielded D
max
of 16 nm
and r
G
of 5.160.1 nm (Fig. 3B, Table 1). Interestingly, a number
of alternative conformations of the SLI/II/III in solution were
observed from the ab initio analysis of SAXS data with identical
D
max
and r
G
values (Fig. 3C). This observation can likely be
attributed to the underlying flexibility of the RNA molecule in
solution. In a number of the determined solution conformations,
three distinct protrusions are observed which may correspond to
SLI (longer arm), SLII, and SLIII (shorter arm). Fig. 3D presents
the averaged model obtained from superimposing the individual ab
initio models calculated for SLI/II/III. Calculated hydrodynamic
properties (r
H
,r
G
and D
max
) of the SLI/II/III were determined
based on the ab initio models, and excellent agreement was found
with the experimentally determined hydrodynamic properties
(Table 1).
The SLI/II/III of WNV interacts with and activates OAS1
in vitro
The SLI/II/III and initial coding region of the WNV genome is
comprised of three double-stranded stem loops. We sought to
determine whether this region of the genome could bind to and
activate OAS1 in vitro. EMSA experiments of the SLI/II/III under
non-denaturing conditions demonstrated that the RNA interacts
with human OAS1 (Fig. 4A). With increasing concentrations of
OAS1, the SLI/II/III is shifted into a higher molecular weight
complex of increasing intensity. Interestingly, unbound SLI/II/III
(Fig. 4A, lane 1) displays heterogeneity under native conditions
consistent with the subspecies observed by SAXS. This heteroge-
neity is not observed under denaturing conditions (data not
shown).
To identify the specific region(s) within the SLI/II/III that
interacts with OAS1, we generated five different RNA molecules
in addition to the SLI/II/III region that represent either two stem
loops in combination (SLI/II, SLII/III) or individual stem loops
(SLI, SLII and SLIII). We believe this truncation approach is
feasible based on the predicted secondary structure and our
observed solution conformation of the SLI/II/III. Complex
formation with increasing OAS1 concentration was then per-
formed for each RNA molecule in order to determine which
secondary structural elements mediated the interaction. We
observed significant complex formation that appeared as a high
molecular weight species in the EMSA of SLI/II with OAS1
(Fig. 4B). At a constant RNA concentration of 100 nM, no
detectable complex formation was observed with any individual
stem loop (SLI, SLII, or SLIII or with the pairwise combination of
SLII/III in the concentration range of 0 to 1 mM of OAS1
(Fig. 4C).
Upon establishing a direct interaction between the SLI/II/III of
WNV and OAS1 in vitro, we were further interested to investigate
whether this interaction leads to activation of OAS1 catalytic
activity. A colorimetric assay was performed which correlates the
detection of pyrophosphate (PPi) with the production of 29-59(A)
chains. We prepared buffered reactions containing OAS1, ATP,
and Mg
2+
in the presence of SLI/II/III, polyinosinic-polycytidylic
acid (poly I:C, a positive control synthetic dsRNA activator of
OAS1), or a single-stranded RNA (ssRNA) negative control. The
experiments were performed over a 120-minute period, followed
by progressive measurement of PPi production (Fig. 5A). As
expected, ssRNA negative control demonstrated no significant
stimulation of OAS1 activity. The SLI/II/III activates OAS1 to a
Regulation of OAS1 by the 59-UTR of the WNV Genome
PLOS ONE | www.plosone.org 4 March 2014 | Volume 9 | Issue 3 | e92545

Figure 2. Recombinant human OAS1 adopts a globular fold. (A) Sedimentation velocity (SV) distribution analysis in terms of c(S) at 0.4 mg/
mL. In-set is the resultant concentration dependence of the SV distribution. (B) Concentration dependence of hydrodynamic radius obtained from
DLS measurements. (C) The pair distribution function versus particle radius obtained from the GNOM analysis. In-set is the merged scattering data
obtained from multiple concentrations. (D) Superimposition of the human OAS1 (PDB 4IG8) high-resolution structure [26] on the ab initio model
generated using DAMMIF on the data obtained from SAXS experiments on human OAS1.
doi:10.1371/journal.pone.0092545.g002
Table 1. Experimental and predicted hydrodynamic parameters of OAS1 and SLI/II/III (error shown in parentheses).
OAS1 SLI/II/III
HYDROPRORO HYDROPRO
Parameter Experimental DAMMIF BUNCH 4IG8
f
Experimental DAMMIF
r
H
(nm)
a
3.0 (0.3) 3.13 (0.02) 3.05 (0.04) 2.90 5.1 (0.2) 5.00 (0.02)
S
u
20,w
(S)
b
3.26 (0.05) 3.13 (0.01) 3.23 (0.02) 3.16 ND ND
r
G
(nm)
c
2.28 (0.02)
d
2.40 (0.01) 2.23 (0.01) 2.22 5.1 (0.1) 5.10 (0.01)
D
max
(nm)
c
7.1
e
6.90 (0.04) 6.80 (0.04) 6.6 16.0 16.8 (0.01)
x- 0.9 1.0 - - 1.0
NSD - 0.52 (0.02) 0.36 (0.03) - - 1.10 (0.06)
a
experimentally determined from DLS data.
b
from AUC-SV data.
c
from SAXS data.
d
the r
G
values for R195E and K199E are 2.43 (0.11) nm and 2.40 (0.13) nm respectively.
e
the D
max
values for R195E and K199E are 6.9 nm and 7.0 nm respectively.
f
based on homology with high-resolution structure of human OAS1.
doi:10.1371/journal.pone.0092545.t001
Regulation of OAS1 by the 59-UTR of the WNV Genome
PLOS ONE | www.plosone.org 5 March 2014 | Volume 9 | Issue 3 | e92545

level that is approximately 60% of that achieved by poly I:C at all
time points observed (on a per mass basis). However, direct
quantitative comparison to the poly I:C control should be treated
with caution given the extremely large size and heterogeneity of
poly I:C (,90 to 1400 kDa) relative to the WNV RNAs examined.
Taken together, the structured SLI/II/III region of the WNV
genome interacts with and activates OAS1 in vitro.
SLI is dispensable for maximal OAS1 activation
To compare the ability of the stem loop regions of the SLI/II/III
to activate OAS1, time courses monitoring of PPi production were
performed and compared with the full-length SLI/II/III, poly I:C
(positive control) and ssRNA (negative control). Remarkably, of all
the SLI/II/III truncations, only SLII/III is capable of achieving
activation level comparable to the full length SLI/II/III (Fig. 5A).
Figure 3. Solution conformations of the WNV SLI/II/III from SAXS. (A) Dynamic light scattering profile of SLI/II/III at 2 mg/mL. (B) Pair
distribution function of SLI/II/III obtained from merged data of multiple concentrations. In-set is the merged SAXS data obtained from multiple
concentrations. (C) Individual ab-initio models calculated from the SAXS data using DAMMIF program demonstrating two distinct subpopulations of
the RNA molecule. (D) Averaged model of SLI/II/III obtained from individual models presented in Fig. 3C.
doi:10.1371/journal.pone.0092545.g003
Regulation of OAS1 by the 59-UTR of the WNV Genome
PLOS ONE | www.plosone.org 6 March 2014 | Volume 9 | Issue 3 | e92545

Basal levels of PPi production, comparable to the negative control,
were observed for SLI/II and each of the individual stem loops (SLI,
SLII, and SLIII). Interestingly, the SLI/II construct that demon-
strated high affinity complex formation with OAS1 did not
stimulate catalytic activity, while SLII/III demonstrated potent
activation despite a lack of detectable complex formation.
Due to the observed discrepancy between binding and
activation, detailed dose-response experiments in which initial
reaction velocities of OAS1-catalyzed PPi production were
performed under increasing RNA concentrations for each of the
SLI/II/III constructs (Fig. 5B). This approach allowed estimation
of both the apparent dissociation constant (K
app
), a measure of
affinity, and the maximum reaction velocity (V
max
), a measure of
catalysis. Table 2 summarizes the determined kinetic parameters
for each RNA molecule, and includes a measure of the quality of
the fit to the data (R
fit
). The SLI/II/III and SLII/III demonstrate
potent stimulation of OAS1 activity (,200-fold enhancement
relative to the ssRNA negative control for both) despite the 4-fold
higher affinity for OAS1 shown by SLI/II/III compared to the
SLII/III. Despite having a V
max
value approaching that of the
negative control, the SLI/II RNA demonstrated a nearly 3-fold
increase in affinity relative to the SLI/II/III. None of the
individual stem loop structures reveals appreciable stimulation of
OAS1 catalytic activity, and only SLI demonstrates detectable
affinity (4-fold lower affinity than the SLI/II/III). Together, the
kinetic analysis supports a model where SLII/III is the minimal
construct capable of OAS1 activation despite having a weak
binding affinity for the enzyme.
Mutations to the dsRNA binding site disrupt activation of
OAS1 by the SLI/II/III
To confirm that OAS1 activation is an RNA-mediated effect,
the binding affinity and catalytic activity of two point mutants of
human OAS1 (R195E and K199E) were investigated. Based on
the human OAS1 structure, these mutations are in the positively
charged dsRNA-binding groove on the enzyme face distal to the
active site [26]. To verify that the mutations did not disrupt the
native protein conformation, we performed SAXS experiments on
OAS1 R195E and K199E. The resultant pair distribution function
plots for wild type and mutants were nearly identical (Fig. 6A),
and the determined r
G
and D
max
values were within error of the
wild type results (data not shown). Therefore, we conclude that
these mutations do not affect the solution conformation of OAS1.
As expected, higher molecular weight RNA-protein complexes
were not observed in EMSAs of SLI/II/III with increasing
concentrations of R195E or K199E OAS1 (Fig. 6B). We next
investigated whether this loss of interaction had a similar impact
on activation of OAS1 catalytic activity in the presence of SLI/II/
III. Time course experiments following 29-59(A) synthesis by the
mutants in the presence of SLI/II/III (Fig. 6C), SLII/III (data
not shown), or poly I:C (data not shown) confirmed the expected
attenuation of catalytic activity. For example, at the 90-minute
time point, the R195E and K199E mutants demonstrated 3% and
0.4% of wild type activity respectively, in the presence of SLI/II/
III. In an attempt to quantitate the impact, initial reaction
velocities for wild type, R195E and K199E OAS1 were
determined in a dsRNA dose response experiment using SLI/
II/III as the activator (Fig. 6D). The low levels of catalytic activity
demonstrated by the R195E and K199E mutants made accurate
parameter determination impossible (Table 3). Therefore, we
Figure 4. The WNV SLI/II/III forms a direct interaction with human OAS1. (A) EMSA for OAS1 (100 nM) binding to the SLI/II/III under non-
denaturing conditions. (B) EMSA for OAS1 (100 nM) binding to SLI+II under non-denaturing conditions. (C) Non-denaturing gel electrophoresis of
SLI/II/III truncations (100 nM) in the presence or absence of OAS1 (400 nM). In all cases, 8% native TBE gels were used and stained with Sybr Gold
(Invitrogen, USA) to visualize RNA-containing species.
doi:10.1371/journal.pone.0092545.g004
Regulation of OAS1 by the 59-UTR of the WNV Genome
PLOS ONE | www.plosone.org 7 March 2014 | Volume 9 | Issue 3 | e92545

conclude that the SLI/II/III of WNV is mediating its effects
through interaction with the previously reported dsRNA-binding
site on OAS1 [26].
Single nucleotide polymorphism in the OAS1 gene does
not impede activation by dsRNA
The S162G mutation has been reported in a previous study as a
very common single nucleotide polymorphism (SNP) in the OAS1
gene, and is more prevalent in WNV-susceptible individuals [55].
We therefore overexpressed and purified the mutant version of the
protein, and investigated its ability to bind to and be activated by
the SLI/II/III of WNV. Overall, no significant differences were
observed for this mutant in terms of solution conformation
(Fig. 6A), affinity for SLI/II/III (Fig. 6B), or ability to perform
catalysis in the presence of dsRNA (Fig. 6C, D). These results are
not surprising as this SNP is located on the face opposite the active
site aspartic acids and does not mediate interactions with dsRNA
[26,55]. A comparison of the determined kinetic parameters is
shown in Table 3.
Discussion
OAS1 and other OAS isoforms play an important role in the
recognition of viral dsRNA and subsequent amplification of the
initial interferon-mediated innate immune response [24,25,27,28].
Previous studies have linked SNPs in the OAS1 gene to
susceptibility to WNV infection [31,32]. The p42 isotype of
OAS1 used our studies has been previously implicated to combat
Dengue virus infection via an RNase L dependent pathway [34].
To best of our knowledge no direct enzymatic activation studies of
OAS1 by regions of the WNV genome been previously performed.
We therefore sought to investigate whether the WNV RNA
genome served as a source for OAS1 activation. Our initial
investigations focused on the SLI/II/III, based on its secondary
structure that is conserved amongst Flaviviridae family members.
We conclude that a direct interaction between the SLI/II/III of
the WNV and OAS1 occurs in vitro, and therefore warrants further
investigation in a cellular context.
Our study found that the SLI/II/III of WNV is a potent
activator of OAS1 in vitro. The affinity of OAS1 for the SLI/II/III
(K
app
of 14536199 nM) is consistent with affinities for other short
viral RNAs [36], and the maximum catalytic activity (V
max
of
2162mM/min) is within error of the positive control (synthetic
poly I:C) (Table 2). Examination of various SLI/II/III trunca-
tions enabled a comprehensive analysis of RNA binding and
activation potential to narrow down sufficient region(s). In tandem
SLII/III are necessary for catalytic activation of OAS1 despite the
requirement of SLI in conjunction with SLII for the highest
affinity interaction. Stem loops SLII/III has a higher K
app
that
shows weaker binding, which is supported by absence of any
higher species in its EMSA in presence of OAS1 but also has a
high V
max
comparable to poly I:C and SLI/II/III of WNV. This
Figure 5. Catalytic activation of OAS1 by the SLI/II/III and its
truncations. (A) Purified OAS1 (300 nM) and RNA (300 nM) were
incubated in the presence of ATP (2 mM) and MgCl
2
(5 mM) at 37uC,
quenched at time points from 0–180 minutes, and 29-59(A) chain
formation quantitated by PP
i
detection. In all cases, errors represent the
standard deviation from at least 3 replicates, and ssRNA represents a
single-stranded negative control. (B) Enzymatic activity of OAS1
(400 nM) shown as a function of RNA concentration. Linear regression
analysis of the initial velocity was used to determine OAS1 activity and
the error in the analysis represented as error bars.
doi:10.1371/journal.pone.0092545.g005
Table 2. Comparison of kinetic parameters (K
app
and V
max
)of
enzymatic activity of wild type OAS1 when activated by WNV
SLI/II/III and its truncations.
RNA
K
app
V
max
R
fit
(nM) (mM/min)
I/II/III 14536199 2162 0.997
I/II 528696 0.960.2 0.994
II/III 57756452 2262 0.995
I 58326387 1.360.3 0.997
II NA NA NA
III NA NA NA
ssRNA 83665 0.160.1 0.347
Poly I:C* 121613 21.660.8 0.995
doi:10.1371/journal.pone.0092545.t002
Regulation of OAS1 by the 59-UTR of the WNV Genome
PLOS ONE | www.plosone.org 8 March 2014 | Volume 9 | Issue 3 | e92545

result supports previous observations that indicate that binding
affinity does not necessarily correlate with the ability of an RNA
molecule to stimulate the synthetase activity of OAS1; instead
balance between sufficient affinity and stimulatory potential is the
key [36,53]. Overall, the SLI/II/III appears to achieve the best
balance, but the construct lacking SLI is capable of achieving a
similar maximum catalytic output to the full length SLI/II/III at
higher RNA concentrations (Table 2).
We studied two point mutations in OAS1 within the RNA
binding domain on the basic tract opposite the active site [26,35].
These mutations did not impact the overall structure of OAS1 as
determined by SAXS experiments (Fig. 6A). As expected, these
mutants show no detectable affinity for WNV SLI/II/III, nor do
Figure 6. Analysis of OAS1 mutants. (A) Pair distribution function versus particle radius obtained from GNOM analysis for wild-type OAS1 (red),
R195E (blue) and K199E (green). Inset, is a SDS-polyacrylamide gel presenting wild type and mutant OAS1s that suggests that all constructs have
similar molecular weight. (B) EMSA for OAS1 and mutants (100 nM) binding to WNV SLI/II/III under non-denaturing conditions. (C) Reactions
containing purified OAS1 or OAS1 mutants (300 nM) and RNA (300 nM) quenched at time points from 0–180 minutes followed by quantification of
PP
i
production. In all cases, errors represent the standard deviation from at least 3 replicates, and ssRNA represents a single-stranded negative
control. (D) Enzymatic activity of OAS1 or OAS1 mutants (400 nM) shown as a function of RNA concentration. Linear regression analysis of the initial
velocity was used to determine OAS1 activity and the error in the analysis represented as error bars.
doi:10.1371/journal.pone.0092545.g006
Regulation of OAS1 by the 59-UTR of the WNV Genome
PLOS ONE | www.plosone.org 9 March 2014 | Volume 9 | Issue 3 | e92545

they activate the catalytic activity of OAS1 (Table 3). This finding
supports a previous study highlighting the importance of R195 and
K199 residues in the proposed dsRNA binding site of OAS1 and
confirms that dsRNA binding via these basic residues is crucial to
the impact of WNV RNA on OAS1 activation [35]. Furthermore,
this result highlights that while the RNA-protein interaction
observed is in the mM range for the SLI/II/III, this affinity is more
than sufficient for a dsRNA activator to activate OAS1.
Genome cyclization has been established as necessary for WNV
genomic RNA replication, and involves a panhandle structure
comprising long-range base pairing interactions between nucleo-
tides in the SLI/II/III (59-UAR in SLII and 59-CS in SLIII and
their complimentary nucleotides in 39-UTR [13,14,17]. For
genome cyclization to occur, both SLII and SLIII in the SLI/
II/III unwind to form their new interactions. RNAse probing
experiments have shown that while SLII and SLIII do adopt these
long-range interactions, SLI of the SLI/II/III remains intact upon
cyclization [11]. Our observations that the RNA construct
comprising SLII/III is sufficient for maximum catalytic activation
of OAS1 is particularly interesting in this context, as the
replication-competent conformation of the SLI/II/III could
potentially evade the OAS1-mediated innate immune response.
Experiments are currently underway to investigate whether the
panhandle structure attenuates activation of OAS1.
Although SAXS represents a low-resolution approach, the
results presented are the first direct structural observation of the
SLI/II/III from the WNV genome. The determined SAXS
models of the SLI/II/III RNA suggest an inherently flexible
molecule in solution. One subset of these conformations presented
three distinct protrusions (that likely correspond to SLI, SLII, and
SLIII respectively), whereas the remaining models lack domain
resolution (Fig. 3C). The averaged solution conformation
(Fig. 3D) has a relatively low NSD value (1.1060.06), suggesting
that these conformations are closely related to each other
structurally. The most straightforward interpretation of these data
is that a dynamic equilibrium between these conformations exists,
and this idea is supported qualitatively by native gel electropho-
resis where at least two distinct RNA conformations are observed.
Given that genome cyclization with complimentary regions in the
39-UTR would require unwinding of both SLII and SLIII in the
SLI/II/III, it is enticing to speculate that our observed confor-
mations represent the ‘‘structured’’ and ‘‘partially unwound’’ SLI/
II/III conformations. As the structural features of the WNV 59-
UTR appear conserved amongst other Flaviviridae family members
at the secondary structural level [11,15,16], the stem-loop
recognition at the 59-end observed in this study by OAS1 may
possibly represent a general feature of OAS enzymes.
Acknowledgments
The authors would like to acknowledge the Manitoba Institute for
Materials for their support and access to infrastructure.
Author Contributions
Conceived and designed the experiments: SD TRP ED EPB KZ KM SEH
SAM. Performed the experiments: SD TRP ED EPB KZ KM SEH.
Analyzed the data: SD TRP SEH SAM. Contributed reagents/materials/
analysis tools: SD TRP ED EPB KZ KM SEH SAM. Wrote the paper: SD
TRP SAM.
References
1. Gubler DJ (2007) The continuing spread of West Nile virus in the western
hemisphere. Clin Infect Dis 45: 1039–1046.
2. Asnis DS, Conetta R, Teixeira AA, Waldman G, Sampson BA (2000) The West
Nile virus outbreak of 1999 in New York: the Flushing Hospital experience. (vol
30, pg 413, 2000). Clinical Infectious Diseases 30: 841–841.
3. Savage HM, Ceianu C, Nicolescu G, Karabatsos N, Lanciotti R, et al. (1999)
Entomologic and avian investigations of an epidemic of West Nile fever in
Romania in 1996, with serologic and molecular characterization of a virus
isolate from mosquitoes. Am J Trop Med Hyg 61: 600–611.
4. Lanciotti RS, Ebel GD, Deubel V, Kerst AJ, Murri S, et al. (2002) Complete
genome sequences and phylogenetic analysis of West Nile virus strains isolated
from the United States, Europe, and the Middle East. Virology 298: 96–105.
5. Calisher CH (2000) West Nile virus in the New World: appearance, persistence,
and adaptation to a new econiche—an opportunity taken. Viral Immunol 13:
411–414.
6. Brinton MA (2001) Host factors involved in West Nile virus replication. West
Nile Virus: Detection, Surveillance, and Control 951: 207–219.
7. Brinton MA (2002) The molecular biology of West Nile virus: A new invader of
the Western hemisphere. Annual Review of Microbiology 56: 371–402.
8. Chambers TJ, Hahn CS, Galler R, Rice CM (1990) Flavivirus Genome
Organization, Expression, and Replication. Annual Review of Microbiology 44:
649–688.
9. Ivanyi-Nagy R, Darlix JL (2012) Core protein-mediated 59-39annealing of the
West Nile virus genomic RNA in vitro. Virus Res 167: 226–235.
10. Markoff L (2003) 59-and39-noncoding regions in flavivirus RNA. Flaviviruses:
Structure, Replication and Evolution 59: 177–228.
11. Zhang B, Dong H, Stein DA, Iversen PL, Shi PY (2008) West Nile virus genome
cyclization and RNA replication require two pairs of long-distance RNA
interactions. Virology 373: 1–13.
12. Borisevich V, Seregin A, Nistler R, Mutabazi D, Yamshchikov V (2006)
Biological properties of chimeric West Nile viruses. Virology 349: 371–381.
13. Villordo SM, Gamarnik AV (2009) Genome cyclization as strategy for flavivirus
RNA replication. Virus Res 139: 230–239.
14. Friebe P, Harris E (2010) Interplay of RNA elements in the dengue virus 59and
39ends required for viral RNA replication. J Virol 84: 6103–6118.
15. Lodeiro MF, Filomatori CV, Gamarnik AV (2009) Structural and functional
studies of the promoter element for dengue virus RNA replication. J Virol 83:
993–1008.
16. Polacek C, Foley JE, Harris E (2009) Conformational changes in the solution
structure of the dengue virus 59end in the presence and absence of the 39
untranslated region. J Virol 83: 1161–1166.
17. Friebe P, Shi PY, Harris E (2011) The 59and 39downstream AUG region
elements are required for mosquito-borne flavivirus RNA replication. J Virol 85:
1900–1905.
18. Kajaste-Rudnitski A, Mashimo T, Frenkiel MP, Guenet JL, Lucas M, et al.
(2006) The 29,59-oligoadenylate synthetase 1b is a potent inhibitor of West Nile
virus replication inside infected cells. J Biol Chem 281: 4624–4637.
19. Samuel CE (2001) Antiviral actions of interferons. Clinical Microbiology
Reviews 14: 778–809.
20. Keller BC, Fredericksen BL, Samuel MA, Mock RE, Mason PW, et al. (2006)
Resistance to alpha/beta interferon is a determinant of West Nile virus
replication fitness and virulence. Journal of Virology 80: 9424–9434.
21. Sadler AJ, Williams BR (2008) Interferon-inducible antiviral effectors. Nat Rev
Immunol 8: 559–568.
22. Peisley A, Hur S (2012) Multi-level regulation of cellular recognition of viral
dsRNA. Cell Mol Life Sci.
23. Zhou A, Hassel BA, Silverman RH (1993) Expression cloning of 2-5A-
dependent RNAase: a uniquely regulated mediator of interferon action. Cell 72:
753–765.
24. Hovanessian AG, Justesen J (2007) The human 29-59oligoadenylate synthetase
family: unique interferon-inducible enzymes catalyzing 29-59instead of 39-59
phosphodiester bond formation. Biochimie 89: 779–788.
Table 3. Comparison of kinetic parameters (K
app
and V
max
)of
enzymatic activity of wild type and mutant OAS1s when
activated by the WNV SLI/II/III.
OAS1
K
app
V
max
R
fit
(nM) (mM/min)
WT 14536199 2162 0.997
S162G 14906188 2262 0.998
R195E NA NA NA
K199E NA NA NA
doi:10.1371/journal.pone.0092545.t003
Regulation of OAS1 by the 59-UTR of the WNV Genome
PLOS ONE | www.plosone.org 10 March 2014 | Volume 9 | Issue 3 | e92545

25. Diamond MS (2009) Mechanisms of evasion of the type I interferon antiviral
response by flaviviruses. J Interferon Cytokine Res 29: 521–530.
26. Donovan J, Dufner M, Korennykh A (2013) Structural basis for cytosolic
double-stranded RNA surveillance by human oligoadenylate synthetase 1. Proc
Natl Acad Sci U S A 110: 1652–1657.
27. Player MR, Torrence PF (1998) The 2-5A system: modulation of viral and
cellular processes through acceleration of RNA degradation. Pharmacol Ther
78: 55–113.
28. Han Y, Whitney G, Donovan J, Korennykh A (2012) Innate Immune Messenger
2-5A Tethers Human RNase L into Active High-Order Complexes. Cell Rep 2:
902–913.
29. Samuel MA, Whitby K, Keller BC, Marri A, Barchet W, et al. (2006) PKR and
RNase L contribute to protection against lethal West Nile Virus infection by
controlling early viral spread in the periphery and replication in neurons. Journal
of Virology 80: 7009–7019.
30. Sangster MY, Urosevic N, Mansfield JP, Mackenzie JS, Shellam GR (1994)
Mapping the Flv locus controlling resistance to flaviviruses on mouse
chromosome 5. J Virol 68: 448–452.
31. Lim JK, Lisco A, McDermott DH, Huynh L, Ward JM, et al. (2009) Genetic
variation in OAS1 is a risk factor for initial infection with West Nile vir us in
man. PLoS Pathog 5: e1000321.
32. Bigham AW, Buckingham KJ, Husain S, Emond MJ, Bofferding KM, et al.
(2011) Host genetic risk factors for West Nile virus infection and disease
progression. PLoS One 6: e24745.
33. Chebath J, Benech P, Revel M, Vigneron M (1987) Constitutive expression of
(29-59) oligo A synthetase confers resistance to picornavirus infection. Nature
330: 587–588.
34. Lin RJ, Yu HP, Chang BL, Tang WC, Liao CL, et al. (2009) Distinct antiviral
roles for human 29,59-oligoadenylate synthetase family members against dengue
virus infection. J Immunol 183: 8035–8043.
35. Hartmann R, Justesen J, Sarkar SN, Sen GC, Yee VC (2003) Crystal structure of
the 29-specific and double-stranded RNA-activated interferon-induced antiviral
protein 29-59-oligoadenylate synthetase. Mol Cell 12: 1173–1185.
36. Meng H, Deo S, Xiong S, Dzananovic E, Donald LJ, et al. (2012) Regulation of
the interferon-inducible 29-59-oligoadenylate synthetases by adenovirus VA(I)
RNA. J Mol Biol 422: 635–649.
37. McKenna SA, Kim I, Puglisi EV, Lindhout DA, Aitken CE, et al. (2007)
Purification and characterization of transcribed RNAs using gel filtration
chromatography. Nat Protoc 2: 3270–3277.
38. Booy EP, Meng H, McKenna SA (2012) Native RNA purification by gel
filtration chromatography. Methods Mol Biol 941: 69–81.
39. Patel TR, Morris GA, de la Torre JG, Ortega A, Mischnick P, et al. (2008)
Molecular flexibility of methylcelluloses of differing degree of substitution by
combined sedimentation and viscosity analysis. Macromol Biosci 8: 1108–1115.
40. Dam J, Schuck P (2004) Calculating sedimentation coefficient distributions by
direct modeling of sedimentation velocity concentration profiles. Numerical
Computer Methods, Pt E 384: 185–212.
41. Schuck P (1998) Sedimentation analysis of noninteracting and self-associating
solutes using numerical solutions to the Lamm equation. Biophysical Journal 75:
1503–1512.
42. Laue TM, Shah BD, Ridgeway TM, Pell etier SL (1992) Computer-aided
interpretation of analytical sedimentation data for proteins. In: Harding SE,
Rowe AJ, Horton JC, editors. Analytical Ultracentrifugation in Biochemistry
and Polymer Science. Cambridge, United Kingdom: Royal Society of
Chemistry. pp. 90–125.
43. Patel TR, Morris GA, Zwolanek D, Keene DR, Li J, et al. (2010) Nano-structure
of the laminin gamma-1 short arm reveals an extended and curved multidomain
assembly. Matrix Biol 29: 565–572.
44. Dzananovic E, Patel TR, Deo S, McEleney K, Stetefeld J, et al. (2013)
Recognition of viral RNA stem-loops by the tandem double-stranded RNA
binding domains of PKR. RNA 19: 333–344.
45. Konarev PV, Volkov VV, Sokolova AV, Koch MHJ, Svergun DI (2003)
PRIMUS: a Windows PC-based system for small-angle scattering data analysis.
Journal of Applied Crystallography 36: 1277–1282.
46. Svergun DI (1992) Determination of the Regularization Parameter in Indirect-
Transform Methods Using Perceptual Criteria. Journal of Applied Crystallog-
raphy 25: 495–503.
47. Franke D, Svergun DI (2009) DAMM IF, a program for rapid ab-initio shape
determination in small-angle scattering. Journal of Applied Crystallography 42:
342–346.
48. Petoukhov MV, Svergun DI (2005) Global rigid body modeling of macromo-
lecular complexes against small-angle scattering data. Biophys J 89: 1237–1250.
49. Patel TR, Reuten R, Xiong S, Meier M, Winzor DJ, et al. (2012) Determination
of a molecular shape for netrin-4 from hydrodynamic and small angle X-ray
scattering measurements. Matrix Biol 31: 135–140.
50. Volkov VV, Svergun DI (2003) Uniqueness of ab initio shape determination in
small-angle scattering. Journal of Applied Crystallography 36: 860–864.
51. de la Torre JG, Huertas ML, Carrasco B (2000) Calculation of hydrodynamic
properties of globular proteins from their atomic-level structure. Biophysical
Journal 78: 719–730.
52. Gasteiger E, Hoogland C, Gattiker A, Duvaud S, Wilkins MR, et al. (2005)
Protein Identification and Analysis Tools on the ExPASy Server. In: Walker JM,
editor. The Proteomics Protocols Handbook: Humana Press. pp. 571–607.
53. Hartmann R, Norby PL, Martensen PM, Jorgensen P, James MC, et al. (1998)
Activation of 29-59oligoadenylate synthetase by single-stranded and double-
stranded RNA aptamers. J Biol Chem 273: 3236–3246.
54. Wathelet M, Moutschen S, Cravador A, DeWit L, Defilippi P, et al. (1986) Full-
length sequence and expression of the 42 kDa 2-5A synthetase induced by
human interferon. FEBS Lett 196: 113–120.
55. Yakub I, Lillibridge KM, Moran A, Gonzalez OY, Belmont J, et al. (2005)
Single nucleotide polymorphisms in genes for 29-59-oligoadenylate synthetase
and RNase L inpatients hospitalized with West Nile virus infection. J Infect Dis
192: 1741–1748.
Regulation of OAS1 by the 59-UTR of the WNV Genome
PLOS ONE | www.plosone.org 11 March 2014 | Volume 9 | Issue 3 | e92545