Aureusvirus P14 Is an Efficient RNA Silencing Suppressor That Binds Double-Stranded RNAs without Size Specificity

Article (PDF Available)inJournal of Virology 79(11):7217-26 · July 2005with53 Reads
DOI: 10.1128/JVI.79.11.7217-7226.2005 · Source: PubMed
RNA silencing is a conserved eukaryotic gene regulatory system in which sequence specificity is determined by small RNAs. Plant RNA silencing also acts as an antiviral mechanism; therefore, viral infection requires expression of a silencing suppressor. The mechanism and the evolution of silencing suppression are still poorly understood. Tombusvirus open reading frame (ORF) 5-encoded P19 is a size-selective double-stranded RNA (dsRNA) binding protein that suppresses silencing by sequestering double-stranded small interfering RNAs (siRNAs), the specificity determinant of the antiviral silencing system. To better understand the evolution of silencing suppression, we characterized the suppressor of the type member of Aureusviruses, the closest relatives of the genus Tombusvirus. We show that the Pothos latent virus (PoLV) ORF 5-encoded P14 is an efficient suppressor of both virus- and transgene-induced silencing. Findings that in vitro P14 binds dsRNAs and double-stranded siRNAs without obvious size selection suggest that P14, unlike P19, can suppress silencing by sequestering both long dsRNA and double-stranded siRNA components of the silencing machinery. Indeed, P14 prevents the accumulation of hairpin transcript-derived siRNAs, indicating that P14 inhibits inverted repeat-induced silencing by binding the long dsRNA precursors of siRNAs. However, viral siRNAs accumulate to high levels in PoLV-infected plants; therefore, P14 might inhibit virus-induced silencing by sequestering double-stranded siRNAs. Finally, sequence analyses suggest that P14 and P19 suppressors diverged from an ancient dsRNA binding suppressor that evolved as a nested protein within the common ancestor of aureusvirus-tombusvirus movement proteins.
JOURNAL OF VIROLOGY, June 2005, p. 7217–7226 Vol. 79, No. 11
0022-538X/05/$08.000 doi:10.1128/JVI.79.11.7217–7226.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Aureusvirus P14 Is an Efficient RNA Silencing Suppressor That Binds
Double-Stranded RNAs without Size Specificity‡
Zsuzsanna Me´rai,
Zolta´n Kere´nyi,
Attila Molna´r,
Endre Barta,
Anna Va´lo´czi,
Gyo¨rgy Bisztray,
Zolta´n Havelda,
Jo´zsef Burgya´n,
and Da´niel Silhavy
Agricultural Biotechnology Center, Go¨do¨llo¨, Hungary,
and Department of Genetics and Horticultural Plant Breeding,
Budapest University of Economic Sciences and Public Administration, Budapest, Hungary
Received 1 September 2004/Accepted 17 January 2005
RNA silencing is a conserved eukaryotic gene regulatory system in which sequence specificity is determined
by small RNAs. Plant RNA silencing also acts as an antiviral mechanism; therefore, viral infection requires
expression of a silencing suppressor. The mechanism and the evolution of silencing suppression are still poorly
understood. Tombusvirus open reading frame (ORF) 5-encoded P19 is a size-selective double-stranded RNA
(dsRNA) binding protein that suppresses silencing by sequestering double-stranded small interfering RNAs
(siRNAs), the specificity determinant of the antiviral silencing system. To better understand the evolution of
silencing suppression, we characterized the suppressor of the type member of Aureusviruses, the closest
relatives of the genus Tombusvirus. We show that the Pothos latent virus (PoLV) ORF 5-encoded P14 is an
efficient suppressor of both virus- and transgene-induced silencing. Findings that in vitro P14 binds dsRNAs
and double-stranded siRNAs without obvious size selection suggest that P14, unlike P19, can suppress
silencing by sequestering both long dsRNA and double-stranded siRNA components of the silencing machin-
ery. Indeed, P14 prevents the accumulation of hairpin transcript-derived siRNAs, indicating that P14 inhibits
inverted repeat-induced silencing by binding the long dsRNA precursors of siRNAs. However, viral siRNAs
accumulate to high levels in PoLV-infected plants; therefore, P14 might inhibit virus-induced silencing by
sequestering double-stranded siRNAs. Finally, sequence analyses suggest that P14 and P19 suppressors
diverged from an ancient dsRNA binding suppressor that evolved as a nested protein within the common
ancestor of aureusvirus-tombusvirus movement proteins.
RNA silencing (also termed posttranscriptional gene silenc-
ing in plants and RNA interference in animals) is a conserved
eukaryotic gene inactivation system that plays regulatory roles
in many biological processes including development, mainte-
nance of genome stability, and antiviral responses (2, 6, 12, 25,
54). RNA silencing is induced by accumulation of double-
stranded RNAs (dsRNAs). dsRNAs are first processed by an
RNase III-like nuclease called DICER (in plants termed
DICER-LIKE, or DCL) into short (21 to 25 nucleotide [nt])
RNAs, and then these short RNAs incorporate and guide
different silencing effector complexes to homologous nucleic
acids for suppression (2, 6, 12, 16, 25, 54). In plants, RNA
silencing acts at both single-cell (cell-autonomous silencing)
and at whole-plant (systemic silencing) levels. Cell-autono-
mous silencing inactivates genes in the cells in which dsRNAs
accumulated. Moreover, cell-autonomous silencing generates
mobile silencing signals that confer suppression of homologous
mRNAs in neighboring cells (short distance) and in distant
tissues (long-distance systemic silencing) (29, 31, 32, 56).
DICERs can process dsRNAs into two functionally different
small RNAs, micro-RNAs (miRNAs) and small interfering
RNAs (siRNAs). miRNAs are involved in the control of many
endogenous protein-encoding mRNAs, while siRNAs mainly
play a role in suppression of molecular parasites such as trans-
posons, transgenes, and viruses (2, 6, 12, 16, 25). In Arabidop-
sis, matured miRNAs are 21- to 22-nt-long single-stranded
RNAs (ssRNAs) that are excised from endogenous hairpin
RNA precursors by DCL1 (60). siRNAs, which are generated
from long dsRNAs, accumulate as short (21 to 22 nt) or long
(23 to 25 nt) double-stranded molecules having 2-nt 3 over-
miRNA- and short siRNA-mediated silencing pathways
share components. Both types of small RNAs are incorporated
into and guide a multicomponent nuclease (RNA-induced si-
lencing complex, or RISC) to homologous mRNAs for sup-
pression. RISC cleaves targeted mRNA in the case of (near)
perfect base-pairing between mRNA and guide RNA. When
the guide RNA is only partially complementary to the mRNA,
RISC mediates translational repression (2, 6, 12). siRNAs also
guide other silencing effector complexes. In addition to RISC,
short siRNAs are supposed to provide sequence specificity for
a host-encoded RNA-dependent RNA polymerase that trans-
forms homologous mRNAs into dsRNAs, thus amplifying si-
lencing. Moreover, short siRNAs could be involved in short-
distance systemic silencing (15, 18). Long siRNAs would play a
role in long-distance systemic silencing (15) and in transcrip-
tional silencing by directing the histone/DNA methylation of
homologous DNA (7, 15, 25, 51).
RNA silencing plays an antiviral role in plants, in insects,
and perhaps in other eukaryotes (2, 13, 22, 37, 46, 59). DCL2
and perhaps other DCL enzymes generate viral siRNAs from
* Corresponding author. Mailing address: Agricultural Biotechnol-
ogy Center, Plant Science Institute, P.O. Box 411, H-2101 Go¨do¨llo¨,
Hungary. Phone: 36 28 526 194. Fax: 36 28 526 145. E-mail: silhavy
† Present address: The Sainsbury Laboratory, John Innes Centre,
Colney Lane, Norwich NR4 7UH, United Kingdom.
‡ Supplemental material for this article may be found at http://jvi
double-stranded replicative intermediates of RNA viruses (61)
or from hairpins of viral mRNAs (48). Viral siRNAs could
target RISC to viral mRNAs for suppression. As RNA-depen-
dent RNA polymerase mutant plants are more susceptible to
certain viruses, it is likely that silencing amplification is also an
important antiviral pathway against particular viruses (9, 11,
30, 61). Importantly, virus-induced silencing acts as a short-
distance systemic defense system. Viral siRNAs might spread
10 to 15 cell layers and activate silencing in still noninvaded
neighboring cells, thus limiting the extent of virus invasion (15,
17, 18, 43).
To counteract RNA silencing, most plant viruses express
silencing suppressor proteins. Viral suppressors target differ-
ent steps of the silencing response (22, 39, 46, 57, 59). Al-
though many suppressors have been identified, the molecular
basis of silencing inhibition and the evolution of suppressors
are poorly understood.
Members of the Tombusviridae plant virus family have ico-
sahedral particles and linear, small, single-stranded positive-
sense RNA genomes (42). Different genera of Tombusviridae
express distinct suppressors. The coat protein (CP) of Turnip
crinkle virus (Tombusviridae, Carmovirus) has multiple func-
tions; in addition to forming a capsid, it also suppresses silenc-
ing (36, 52, 64). By contrast, the 19-kDa suppressor protein
(P19) of tombusviruses (Tombusviridae, Tombusvirus) is appar-
ently required only for silencing inhibition, as it is dispensable
for replication, movement, or virion formation (34, 35, 45, 57).
P19 is a specific dsRNA binding protein, which binds dsRNAs
size selectively (55, 62). P19 forms strong complexes with
dsRNAs having 19-nt duplex regions, thus it binds siRNAs in
vitro (45, 55, 62) and in vivo (8, 14, 21). Importantly, it binds
shorter or longer dsRNAs with much weaker affinity. It is
proposed that in tombusvirus-infected cells, P19 sequesters
silencing-generated siRNAs, thereby suppressing antiviral si-
lencing responses (21, 45). Indeed, in Cymbidium ringspot virus
(CymRSV; Tombusviridae, Tombusvirus)-infected plants, viral
siRNAs are present in complex with P19 (21). To better un-
derstand the evolution of silencing suppression within Tom-
busviridae, we wanted to identify and analyze the silencing
suppressor of Pothos latent virus (PoLV), the type species of
aureusviruses (Tombusviridae). The genome organization of
aureusviruses is identical to that of tombusviruses, but the
Aureusvirus genome is significantly smaller and the sequence
similarity between the two genera is limited (26, 28, 41) (Fig.
1A). The PoLV open reading frame (ORF) 5-encoded 14-kDa
protein (P14), like the tombusvirus ORF 5-encoded P19, in-
creases the severity of viral symptoms (symptom determinant)
(40). Infection of Nicotiana benthamiana plants with either
PoLV14, a mutant PoLV that fails to express P14 (40) (Fig.
1B), or a mutant CymRSV that is unable to express P19
(Cym19stop) leads to similar recovery phenotypes (48). As the
recovery phenotype is supposed to be the manifestation of
virus-induced systemic silencing (1, 17, 37, 48), it has been
suggested that P14, like P19, operates as a silencing suppressor
(38, 40). Interestingly, although the genomic positions of P14
and P19 are identical and both proteins are symptom determi-
nant, no significant sequence homology has been detected be-
tween P19 and P14 (40).
Here we report that PoLV P14 is an efficient suppressor of
both virus- and transgene-induced silencing. P14 is a dsRNA
binding protein that binds dsRNA in vitro without obvious size
selection. The potential mechanism of P14-mediated silencing
suppression and the evolution of P14 and P19 suppressors will
be discussed.
Plant materials and Agrobacterium tumefaciens infiltration. Transgenic N.
benthamiana carrying the green fluorescent protein (GFP) ORF was described
previously (4). The A. tumefaciens infiltration method was carried out as de-
scribed previously (56). For coinfiltration, equal volumes of respective A. tume-
faciens cultures (optical density at 600 nm, 0.25) were mixed before infiltration.
Silencing suppression assay and GFP imaging. The RNA silencing suppres-
sion assay was carried out as described previously (58). Visual detection of GFP
fluorescence was performed using a 100-W handhold long-wave UV lamp (Black
Ray model B 100AP; UV Products, Upland, CA).
Plasmid constructs. The infectious cDNA clones of PoLV, PoLV14 (40),
CymRSV (10), and Cym19stop (48) were described previously.
Silencing suppressors for agroinfiltration assays were cloned into pBIN61S
(45). P19 and Sigma3 binary constructs were described previously (24, 45). P14
was PCR amplified with P14 5 and 3 primers corresponding to the first and last
20 nt of P14. The P14 5 primer carried an additional BamHI site, while the P14
3 primer contained an additional SalI site. The PCR product was cloned in
reverse orientation into SmaI-cleaved pBluescript KS vector (KS-P14). To create
the P14 binary construct, the BamHI-SalI fragment was cloned from KS-P14 into
pBin61S. G-P14 was generated by cloning the BamHI fragment from KS-P14
into Gex-2T, and then the sense orientation was selected. PVX-P14 was gener-
ated by cloning a refilled BamHI-SalI fragment from KS-P14 into EcoRV-
digested pP2C2S.
In vitro RNA transcription and plant inoculation. In vitro transcription from
PoLV, PoLV14 (40), CymRSV, Cym19stop, and PVX cDNA clones and inoc-
ulation of RNA transcripts onto plants were performed as described previously
(10, 48).
FIG. 1. PoLV P14 is a symptom determinant. (A) Schematic rep-
resentation of genome organization of a tombusvirus (CymRSV) and
PoLV, the type species of aureusviruses. K, kilodalton. (B) PoLV P14
increases symptom severity. N. benthamiana plants were infected with
PoLV and PoLV14, a mutant virus that fails to express P14.
PoLV14 infection results in a recovery phenotype characterized by
low virus titers and mild symptoms in the upper leaves. Plants were
grown at 21°C.
7218 ME
Protoplast preparation and inoculation. Protoplasts were isolated from N.
benthamiana and transfected with in vitro transcripts of PoLV or PoLV14 (10,
Protein separation and Western analysis. Proteins were separated in a 12%
sodium dodecyl sulfate-polyacrylamide gel and than transferred onto a Hybond
C Extra filter (Amersham Pharmacia Biotech). Rabbit P14 polyclonal antibodies
(anti-P14) raised against the GST-P14 fusion protein were used for Western
RNA gel blot analysis. The same total RNA extract was used for high- and
low-molecular-weight RNA gel blot analysis. RNA extraction and RNA gel blot
analysis were carried out as described previously (48). PCR fragments labeled
with the random priming method were used for Northern analyses of high-
molecular-weight RNAs. Radioactively labeled in vitro transcripts corresponding
to the positive strand of virus RNA and antisense strand of GFP were used as
probes for Northern analyses of low-molecular-weight RNAs. Labeling was car-
ried out as described previously (48).
Gel mobility shift assay. Synthetic siRNAs were labeled with T4 PNK.
P]UTP-labeled in vitro RNA transcripts were used as long RNA probes.
Transcripts were produced from a T7-T3 Bluescript PCR fragment by the T7 and
T3 RNA polymerases, respectively (48).
To generate double-stranded siRNAs, 5-phosphorylated complementary
strand siRNAs in 5 times molar excesses were added to labeled single-stranded
siRNAs, and then siRNAs were heated and annealed. To generate long dsRNAs,
a 1:1 mixture of labeled T7 and T3 in vitro transcripts were heated and annealed.
GST, G-P14, and G-P19 proteins were expressed and purified according to the
manufacturer’s protocols (Amersham Pharmacia Biotech).
To prepare protein extract, 0.25 g leaf tissue was grinded in 1 ml band shift
buffer (83 mM Tris-HCl [pH 7.5], 0.8 mM MgCl
, 66 mM KCl, 100 mM NaCl,
and 10 mM dithiothreitol), and then this crude extract was centrifuged twice for
15 min at 15,000 g. The supernatant was frozen in aliquots at 70°C. In a
binding reaction, labeled dsRNA (ina1nMconcentration) was incubated with
extract containing 2 g total protein. Binding reaction and mobility shift assays
were carried out as described previously (48), except that 8 U RNasin was added
to each 10-l reaction mixture. For long dsRNA direct competition assays,
0.02% Tween 20 was added to the binding buffer.
Computer analysis. Multiple alignments of RNA and deduced protein sequences
were carried out with ClustalX (53). Relationships among proteins were analyzed by
the bootstrap parsimony (47) and maximum-likelihood methods (44).
P14 is a silencing suppressor. Expression of a silencing
suppressor from a heterologous virus intensifies viral symp-
toms (57). To test whether P14 is a silencing suppressor, we
infected N. benthamiana and Nicotiana clevelandii plants with
Potato virus X (PVX) that expressed P14 (PVXP14) and with
PVX as a control. While PVX infection caused only mild
symptoms on both hosts, PVXP14-infected plants showed
strong symptoms including stunting and necrosis along the
veins (Fig. 2A and data not shown). Findings that P14 in-
creased PoLV and PVX symptoms suggest that P14 is a sup-
pressor of virus-induced silencing.
Viral silencing suppressors can be identified in sense trans-
gene-induced silencing assays (58). Infiltration of the leaves of
an N. benthamiana plant with an Agrobacterium carrying a
plasmid that expresses GFP (35SGFP) leads to strong, tran-
sient GFP expression, but it also triggers GFP silencing (4, 58).
Cell-autonomous GFP silencing is manifest as a weakening of
green fluorescence, a decrease in the level of GFP mRNA, and
an accumulation of both short (21 to 22 nt) and long (23 to 25
nt) GFP-specific siRNAs in the infiltrated patches (Fig. 2B and
C). However, GFP silencing is partially or fully inhibited if
35SGFP is coinfiltrated with a second Agrobacterium express-
ing a silencing suppressor. To determine whether P14 sup-
presses sense transgene-induced silencing, N. benthamiana
plants were coinfiltrated with 35SGFP and with a second
Agrobacterium expressing P14 (P14). As the green fluorescence
was much stronger and lasted longer in coinfiltrated patches
than in leaves infiltrated with 35SGFP alone (Fig. 2B), we
concluded that P14 inhibited sense transgene-induced cell-
autonomous silencing.
The effect of silencing suppressors on accumulation of short
and long siRNAs depends on the targeted step of a particular
suppressor (15, 18). To investigate which step of silencing is
targeted by P14, we studied the accumulation of GFP mRNA
and the GFP-derived siRNAs in 35SGFP- and P14-coinfil-
trated leaves. As Fig. 2C shows, in coinfiltrated leaves, GFP
mRNAs accumulated to much higher levels than in leaves that
FIG. 2. P14 is an RNA silencing suppressor. (A) P14 increases the symptoms of PVX. N. benthamiana plants were infected with PVX or
PVXP14, a modified PVX that expressed P14. (B) P14 suppresses sense transgene-induced RNA silencing. Leaves of N. benthamiana plants were
infiltrated with an Agrobacterium (35SGFP) expressing GFP () or were coinfiltrated with 35SGFP and a second Agrobacterium expressing P14
(P14). Photos were taken at 6 d.p.i. (C) Effect of P14 on accumulation of GFP mRNAs and GFP-derived siRNAs (siRNS).
were infiltrated with 35SGFP alone. Moreover, neither short
nor long GFP-specific siRNAs could be detected in coinfil-
trated leaves. These data indicate that P14 interferes with
sense transgene-induced cell-autonomous silencing by prevent-
ing the accumulation of siRNAs. The effects of a silencing
suppressor on transgene-induced short- and long-distance sys-
temic silencing can be also studied in coinfiltration assays. If a
GFP-transgenic N. benthamiana is infiltrated with 35SGFP,
cell-autonomous GFP silencing generates mobile signals,
which lead to systemic GFP silencing. Since P14 prevents the
accumulation of either short or long siRNAs in coinfiltrated
patches and siRNAs are supposed to play role in systemic
silencing, we postulated that P14 also inhibits the development
of systemic silencing. Indeed, coinfiltration of P14 prevented
the development of both short- and long-distance systemic
GFP silencing (data not shown).
Taken together, the findings that P14 increases viral symp-
toms and inhibits silencing in agroinfiltration assays indicate
that P14 is an efficient suppressor of both virus and transgene-
induced silencing.
P14 is a dsRNA binding protein. In a GFP coinfiltration
(sense transgene-induced silencing) assay, P14 acts like the P19
suppressor of closely related tombusviruses (18, 34, 35, 45).
Both proteins prevent the accumulation of GFP-specific short
and long siRNAs in the coinfiltrated patches, thus inhibiting
the development of cell-autonomous and systemic silencing.
These data open up the possibility that P14 and P19 suppres-
sors target an identical step in the silencing pathway. P19
inhibits silencing by sequestering double-stranded siRNAs. To
test whether P14 could also suppress silencing by binding dou-
ble-stranded siRNAs, we studied the RNA binding activity of
P14 in gel mobility shift assays. P14 was expressed and purified
as a GST fusion protein (G-P14), and then G-P14 was probed
with labeled, synthetic single-stranded and double-stranded
siRNAs. A GST fusion version of the previously characterized
Carnation Italian ringspot tombusvirus (CIRV) P19 (G-P19)
was used as a control (55). As expected, G-P19 did not shift
ssRNAs, while it bound 21-nt double-stranded siRNAs. Im-
portantly, G-P14 also failed to form complexes with ssRNAs
but bound 21-nt double-stranded siRNA (Fig. 3A). These data
suggest that G-P14, like G-P19, is a double-stranded siRNA
binding protein. However, G-P14 forms complexes with dou-
ble-stranded siRNAs with less efficiency than G-P19, since a
similar shift required a much higher G-P14 concentration. The
striking feature of P19-mediated dsRNA binding is its strong
size selectivity. To test whether P14 is also a size-selective
dsRNA binding protein, G-P14 was also probed with long (144
nt) dsRNAs. Interestingly, we found that unlike G-P19, G-P14
formed complexes with long dsRNAs (Fig. 3A). These data
suggest that G-P14 is a dsRNA binding protein that lacks size
specificity. Unfortunately, we failed to release functional P14
from G-P14 with thrombin cleavage; therefore, we cannot
characterize the RNA binding activity of Escherichia coli-ex-
pressed native P14.
Although in vitro G-P14 fusion protein inefficiently forms
complexes with double-stranded siRNAs, it is possible that
native P14 efficiently binds double-stranded siRNAs in PoLV-
infected cells. Moreover, full double-stranded siRNA binding
activity of P14 might require plant-specific posttranslational
modification and/or the presence of host/viral factors. There-
fore we wanted to directly analyze the double-stranded siRNA
binding activity of PoLV-expressed P14. To study whether
PoLV-expressed P14 binds double-stranded siRNAs, we com-
pared the double-stranded siRNA binding activity of crude
extracts prepared from PoLV (PoLV extract)- and PoLV14
(PoLV14 extract)-inoculated N. benthamiana leaves. Impor-
tantly, PoLV extract efficiently bound 21-nt double-stranded
siRNAs, while PoLV14 extract failed to form complexes with
21-nt double-stranded siRNAs (Fig. 3B). As P14 protein was
expressed only in PoLV-infected leaves (see Materials S1 and
Fig. S1A in the supplemental material), it is likely that P14
provided the double-stranded siRNA binding capacity for the
PoLV extract. Moreover, PoLV extract causes almost as strong
a shift on 21-nt double-stranded siRNAs as the extract that was
prepared from CymRSV tombusvirus-inoculated leaves
(CymRSV extract) (Fig. 3B). This result suggests that PoLV-
expressed P14 effectively forms complexes with double-
stranded siRNA.
To investigate whether P14 can bind double-stranded
siRNAs in the absence of viral factors, we analyzed the double-
stranded siRNA binding activity of crude extract prepared
from P14-agroinfiltrated N. benthamiana leaves (P14 extract).
Extract prepared from 35SGFP-infiltrated leaves was used as a
negative control (GFP extract). We could not detect double-
stranded siRNA binding activity in the GFP extract (data not
shown), while P14 extract caused a strong shift on 21-nt dou-
ble-stranded siRNAs (Fig. 3D, second lane). Moreover, as
both GFP and P14 extracts weakly and identically bound sin-
gle-stranded siRNAs (data not shown), it is likely that P14 does
not bind ssRNAs. Therefore, we conclude that in planta ex-
pressed P14 is a double-stranded siRNA binding protein which
does not require viral factors for dsRNA binding.
In vitro, P19 binds dsRNAs size selectively, while G-P14
binds dsRNAs without size specificity. To test whether in
planta-expressed P14 and P19 proteins also differ in dsRNA
preference, we defined the relative affinity of plant-produced
P14 and P19 for 21-nt and 26-nt double-stranded siRNAs. To
aim this, direct competition assays were carried out with P14
and P19 extracts prepared from agroinfiltrated leaves. Labeled
21-nt double-stranded siRNAs were incubated with plant ex-
tracts and with increasing molar concentrations of cold 21-nt
and 26-nt double-stranded siRNA competitors (Fig. 3D and
F). Competition experiments were repeated with labeled 26-nt
double-stranded siRNAs (Fig. 3E and G). In line with results
obtained with P19 expressed in E. coli, the P19 extract showed
a much higher affinity for 21-nt double-stranded siRNAs than
for 26-nt double-stranded siRNAs (55, 62). A large molar
excess of 26-nt double-stranded siRNAs (320) was required
for detectable competition when 21-nt double-stranded siRNA
was labeled (Fig. 3F), while 21-nt double-stranded siRNAs
outcompeted 26-nt double-stranded siRNAs even at a low
(20) molar excess (Fig. 3G). By contrast, the P14 extract had
no obvious size specificity because approximately the same
molar excess of cold 21-nt double-stranded siRNAs and 26-nt
double-stranded siRNAs was required for outcompeting either
labeled 21- or 26-nt double-stranded siRNAs (Fig. 3D and E).
These results suggest that both suppressors bind double-
stranded siRNAs efficiently but with different selectivities. P19
might be more specific for small double-stranded siRNAs,
7220 ME
whereas P14 could bind short and long double-stranded
siRNAs with comparable affinity.
In vitro G-P14 fusion protein binds long dsRNAs. To test
whether in planta-expressed P14 also binds long dsRNAs, la-
beled 144-nt dsRNA was incubated with PoLV and with P14
extract. As negative controls, PoLV14 and GFP extract were
used, respectively. As Fig. 3C shows, a long dsRNA binding
activity could be detected in both PoLV and P14 extracts that
was lacking in either the PoLV14 or GFP extract. Therefore,
it is very likely that P14 provided the long dsRNA binding
activity for PoLV and for P14 extracts. Findings that long
dsRNA binding of P14 extract could be outcompeted with cold
dsRNA but not with ssRNA further support the notion that in
planta-expressed P14 is a dsRNA binding protein (see Mate-
rials S2 and Fig. S1B in the supplemental material). Moreover,
in line with results obtained with G-P19 fusion protein, labeled
long dsRNA was not bound by CymRSV-expressed P19 (Fig.
Collectively, mobility shift assays revealed that P14 and P19
suppressors are different dsRNA binding proteins: P19 is a
strict size-specific dsRNA binding protein, while P14 binds
dsRNAs without strong size preference.
P14 and P19 act differently in hairpin-induced agroinfiltra-
tion assays. As dsRNAs play a key role in silencing and P14 is
a dsRNA binding protein, we postulated that P14-mediated
silencing is based on sequestering a dsRNA component of the
silencing machinery. P14 might sequester long dsRNAs, the
inducers of silencing machinery, or double-stranded siRNAs,
FIG. 3. P14 is a dsRNA binding protein. (A) RNA binding activity of P14 (G-P14) and P19 (G-P19) expressed as a GST fusion protein were
studied in gel mobility shift assays. Shift assays with expressed GST or without additional protein () were used as negative controls. Synthetic
21-nt RNAs were labeled and used as ssRNA probes, while labeled 21-nt RNAs, which annealed to dsRNAs having a 19-nt duplex with 2-nt 3
overhangs (21 ds siRNA) were used as double-stranded siRNA probes; 144-nt-long, labeled, complementary in vitro transcripts were annealed and
used as long dsRNA probes (144 dsRNA). Concentrations of G-P14 of 15 M and 1.5 M were used for the double-stranded siRNA and for the
long dsRNA binding experiments, respectively. A 1.5 M concentration of G-P19 was used for both double-stranded siRNA and long dsRNA
binding tests. Free probes and protein-probe complexes are referred as F and C, respectively. (B) PoLV-expressed P14 efficiently binds
double-stranded siRNAs. dsRNA binding activity of extracts prepared from PoLV- or PoLV14-inoculated leaves were probed with labeled 21-nt
double-stranded siRNAs. Extracts prepared from leaves inoculated with CymRSV and Cym19stop (C19stop), a mutant CymRSV that was unable
to express P19, were used as controls. Extracts were isolated at 3 d.p.i. from N. benthamiana leaves inoculated with the corresponding viruses. (C) In
planta-expressed P14 binds long dsRNA. Extracts prepared from mock ()-, PoLV-, PoLV14-, CymRSV-, and C19stop-inoculated N. benthami-
ana leaves (left panel) were probed with labeled 144-nt dsRNA. Extracts prepared from mock ()-, 35SGFP (GFP)-, and P14-infiltrated N.
benthamiana leaves (right panel) were also probed with labeled 144-nt dsRNA. Each extract was obtained at 3 d.p.i. An asterisk indicates a
nonspecific binding activity that is not associated with P14 or P19 suppressors because it is present in all extracts prepared from PoLV14,
CymRSV, and C19stop virus-inoculated leaves. Similar nonspecific long dsRNA binding activity can be occasionally detected in extracts prepared
from mock- or 35SGFP-infiltrated leaves. (D to G) Plant-expressed P14 binds double-stranded siRNAs without size selectivity. P14 and P19
extracts were obtained from P14- and P19-infiltrated N. benthamiana leaves at 3 d.p.i. Direct competition assays were carried out with labeled 21-nt
(D and F) and 26-nt double-stranded siRNAs (E and G) and with cold competitors added in the indicated molar excesses. A 0 indicates that the
shift assay was conducted in the absence of competitor.
the specificity determinants of the silencing system. To distin-
guish between these two possibilities, we studied the suppres-
sor activity of P14 in hairpin transcript-induced silencing as-
Agroinfiltration with an inverted repeat GFP construct
(GFP IR) leads to expression of hairpin GFP RNAs (GFP-ir).
As hairpin transcripts are rapidly processed into siRNAs by the
silencing machinery, GFP-ir transcripts are barely detectable,
while GFP-ir-derived siRNAs accumulate to high levels in the
infiltrated leaves (Fig. 4C). Moreover, coinfiltration of GFP IR
with 35SGFP (GFP IR plus 35SGFP) prevents transient GFP
activity (Fig. 4A and B) because GFP-ir-derived siRNAs direct
early degradation of GFP mRNAs (19). However, coinfiltra-
tion of GFP IR plus 35SGFP with dsRNA binding proteins
such as reovirus Sigma3 (24) or P19 (50) result in strong green
fluorescence and accumulation of GFP mRNAs (Fig. 4A and
B). Sigma3 and P19 suppress hairpin-induced silencing at dif-
ferent steps. Sigma3 is a strong dsRNA binding protein that
forms complexes only with dsRNAs longer than 30 nt (63).
Sigma3 is proposed to suppress silencing by sequestering hair-
pin transcripts (24) because, in coinfiltrated leaves, GFP-ir
transcripts accumulate to high levels, while siRNAs could not
be detected (Fig. 4B). By contrast, in P19-coinfiltrated leaves
GFP-ir transcripts could not be detected, while siRNAs are
easily detected (50) (Fig. 4B). These data are interpreted to
mean that the siRNA-specific dsRNA binding protein P19
inhibits hairpin-induced silencing by sequestering double-
stranded siRNAs (50).
To test whether P14 can suppress hairpin-induced silencing,
we coinfiltrated leaves with GFP IR plus 35SGFP and with
P14. We found that GFP mRNAs accumulated to high levels
and green fluorescence was strong in P14 coinfiltrated leaves,
indicating that P14 suppressed hairpin-induced silencing effi-
ciently (Fig. 4A). Surprisingly, we failed to detect either hair-
pin transcripts or siRNAs in P14-coinfiltrated leaves (Fig. 4B).
To prove that P14 directly affects on hairpin-derived siRNA
accumulation, we infiltrated leaves with GFP IR or coinfil-
trated leaves with GFP IR and P14. As expected, in GFP
IR-infiltrated samples, siRNA accumulated to high levels,
while hairpin transcripts could not be detected. By contrast, in
GFP IR- and P14-coinfiltrated leaves, neither hairpin tran-
scripts nor siRNAs could be found (Fig. 4C).
Collectively, GFP IR coinfiltration studies revealed that
P14-mediated suppression of hairpin-induced silencing is
mechanistically different than either Sigma3- or P19-mediated
silencing inhibition. We suggest that these differences are the
consequences of the different dsRNA binding preferences of
the Sigma3, P19, and P14 proteins (see Discussion).
P14 fails to prevent accumulation of viral siRNAs. The ob-
servation in agroinfiltration assays that P14 prevents the accu-
mulation of hairpin transcript-derived siRNAs suggests that
P14 suppresses virus-induced silencing by preventing the accu-
mulation of viral siRNAs. To test this model, we monitored the
accumulation of viral RNAs and siRNAs in PoLV- and
PoLV14-inoculated N. benthamiana leaves. By 1 day postin-
oculation (d.p.i.), PoLV and PoLV14 viral RNAs accumu-
lated to detectable levels, while by 2 d.p.i., PoLV and
PoLV14 RNAs were abundant in the inoculated leaves (Fig.
5A). Surprisingly, virus-specific siRNAs could be identified in
both PoLV- and PoLV14-inoculated leaves (Fig. 5A). As the
P14 protein could already be detected at 1 d.p.i. in PoLV-
infected leaves (Fig. 5A bottom panel), we concluded that P14
fails to prevent the accumulation of viral siRNAs. However,
viral siRNA/viral genomic RNA ratios were higher in
PoLV14-inoculated leaves than in PoLV-inoculated ones
(Fig. 5A). These data indicate that virus-induced silencing op-
erated less efficiently in PoLV-inoculated leaves than in
PoLV14-infected tissues, confirming that P14 acts as an effi-
cient suppressor of aureusvirus-induced silencing.
Suppressors of aureusviruses and tombusviruses derive
from a common ancestor protein. Findings that both P14 and
P19 proteins bind double-stranded siRNAs and suppress si-
lencing suggest that these proteins evolved from a common
ancestor. However, the nonrelated P21 suppressor of Beet yel-
lows virus (Closteroviridae, Closterovirus) also binds double-
stranded siRNAs (8), indicating that dsRNA binding silencing
suppressors evolved more than once. Therefore, it is also pos-
sible that P14 and P19 suppressors evolved independently.
To clarify whether P14 and P19 proteins evolved indepen-
dently or have a common ancestor, multiple-sequence align-
ments were carried out with many tombusvirus and aureusvirus
sequences. In both genera, ORF 5 is completely nested within
FIG. 4. P14 suppresses hairpin-induced RNA silencing. (A) N.
benthamiana leaves were infiltrated with 35SGFP and a second
Agrobacterium (GFP IR) expressing hairpin GFP transcripts ()or
coinfiltrated with GFP IR plus 35SGFP and P14, P19, and Sigma3
suppressors, respectively. Photos were taken at 3 d.p.i. (B) Effect of
P14 on accumulation of hairpin transcripts, GFP mRNAs, and siR-
NAs. GFP-ir indicates hairpin transcripts derived from GFP IR, while
GFP refers to GFP mRNA transcribed from 35SGFP (upper panel).
RNA samples were isolated at 3 d.p.i. from GFP IR plus 35SGFP-
infiltrated () or from GFP IR plus 35SGFP and dsRNA binding
protein-coinfiltrated leaves. Note that the probe we used for siRNA
hybridization (bottom panel) detected both hairpin transcript and GFP
mRNA-derived siRNAs (GFP siRNA). (C) P14 prevents the genera-
tion of siRNAs from hairpin transcripts. RNA samples were isolated at
3 d.p.i. from GFP IR-infiltrated () or from GFP IR plus dsRNA
binding protein-coinfiltrated leaves.
7222 ME
ORF 4 but it is translated in the third frame related to ORF 4
(Fig. 1A). Therefore, we first defined the biologically mean-
ingful gaps that optimized both the alignment of ORF 4-en-
coded movement proteins (MP) and the alignment of suppres-
sor proteins (for details regarding defining gaps and creating
protein alignments, see Materials S4 and Fig. S2 in the sup-
plemental material).
Multiple-sequence alignments of proteins deduced from
tombusvirus and aureusvirus ORF 4 (see Materials S4 and Fig.
S3 in the supplemental material) and ORF 5 (Fig. 6) RNA
sequences revealed that similarity was strong for both move-
ment and suppressor proteins within a genus but weak between
aureusviruses and tombusviruses (see Fig. S4 in the supple-
mental material). Amino acids that were identical or similar
between aureusviruses and tombusviruses could be found all
along the MPs except in the C-terminal region (see Materials
S4 and Fig. S3 in the supplemental material). By contrast,
conserved amino acid positions were limited to short regions in
the suppressor alignment (Fig. 6). Interestingly, the regions of
similarity in the suppressor sequences coincide with previously
identified secondary structural elements (1, 2, 3, 4, and
the last -helix) that play key roles in double-stranded siRNA
binding by P19. 4 and the last -helix contribute to the P19
homodimer formation, while the four -strands form a concave
sheet that makes contact with the sugar-phosphate backbone
of double-stranded siRNA (3, 55, 62). No similarity was found
between aureusvirus and tombusvirus suppressors in the region
corresponding to the P19 2 helix (reading head) that interacts
with the end of siRNAs (3, 55, 62) (Fig. 6).
As tombusvirus and aureusvirus suppressors show similarity
in conserved regions that are important for P19-mediated sup-
pression, we suggest that these proteins have evolved from an
ancient suppressor, which was nested in the common ancestor
of the MPs of tombusviruses and aureusvirures. Tombusvirus
and aureusvirus MPs belong to the 30-kDa MP superfamily
(27). To explore the evolution of tombusvirus-aureusvirus sup-
pressors, we sought to determine whether related MPs also
encoded a nested suppressor. The closest relatives to tombus-
virus and aureusvirus MPs are the umbro, tobra (27), and
trichovirus MPs (E. Barta, unpublished data). Importantly,
none of these MP encodes a nested protein (data not shown).
Previously P14 was identified as a symptom determinant
(40). Here we demonstrate that P14 is a dsRNA binding pro-
tein that inhibits virus- and transgene-induced silencing.
P14 is a dsRNA binding protein. P14 is a dsRNA binding
protein which forms complexes with 21- or 26-nt double-
stranded siRNAs and with long dsRNAs (Fig. 3). It was an
unexpected finding because the related P19 preferentially
bound 21-nt double-stranded siRNAs (55, 62). Sequence align-
ments might explain the molecular basis of different double-
stranded siRNA binding of these two suppressors. Regions
that play roles in forming the dsRNA binding surface are
conserved between aureusviruses and tombusviruses, while the
reading head, which is involved in size-specific binding of P19,
cannot be identified in aureusviruses (Fig. 6). Therefore, it is
possible that P14 and P19 suppressors bind double-stranded
siRNAs with similar structure, except that P14 does not inter-
act with the 5 ends of RNAs.
P14-mediated silencing suppression. Previous results have
shown that dsRNA binding proteins could act as silencing
suppressors. P19 and P21 suppressors bind double-stranded
siRNAs in vivo (8, 14, 21), indicating that these proteins inhibit
silencing by sequestering double-stranded siRNAs. It is pro-
posed that the influenza Ns1 protein also inactivates silencing
in plant and insect cells by binding double-stranded siRNAs (5,
FIG. 5. P14 does not prevent accumulation of PoLV-derived
siRNAs. (A) Accumulation of high- and low-molecular-weight viral
RNAs (viral siRNA) in the inoculated leaves of PoLV- and PoLV14-
infected N. benthamiana plants. G indicates genomic virus RNAs,
while Sg1 and Sg2 refer to subgenomic 1 and subgenomic 2 RNAs,
respectively. Note that P14 was already detectable at 1 d.p.i. and that
the level of PoLV Sg2 declines. Numbers below the bottom panel refer
to the ratios of viral siRNA/viral genomic RNAs. The sample loaded in
the first lane was taken as 100% (1), and others were normalized to it.
-P14 refers to polyclonal antibody raised against P14. (B) Virus-
induced cell-autonomous silencing fails to control accumulation of
suppressorless PoLV. RNA samples were isolated from N. benthami-
ana protoplasts transfected with in vitro transcripts of PoLV or
PoLV14. A Western blot assay with the available polyclonal antibody
failed to detect P14 protein in PoLV-transfected cells, therefore con-
clusions about the role of P14 in suppression of virus induced cell-
autonomous silencing cannot be drawn (see Materials S3 in the sup-
plemental material).
23). Moreover, as Sigma3 suppresses transgene-induced silenc-
ing efficiently, it is likely that binding to long dsRNAs could
also inhibit certain silencing pathways (24). Although we can-
not exclude that P14-mediated silencing suppression depends
on its interaction with a host protein, the simplest explanation
is that P14 targets silencing by binding long dsRNAs and/or
double-stranded siRNAs. We suggest that P14 inhibits hairpin-
induced silencing by binding long dsRNAs, while it suppresses
virus-induced silencing by sequestering double-stranded
siRNAs or by delaying the generation of viral siRNAs.
Coinfiltration of P14 with GFP IR plus 35SGFP prevents the
accumulation of both hairpin transcript- and hairpin-derived
siRNA (Fig. 4). These findings can be explained if P14 does not
interfere with the generation of siRNAs, instead it accelerates
the degradation of them. However, viral siRNAs accumulate to
high levels in PoLV-infected cells (Fig. 5). Therefore, we pre-
fer the alternative explanation that P14 binds hairpin tran-
scripts (long dsRNAs); thus, it inhibits DCL-mediated process-
ing of hairpin transcripts but allows their degradation by
alternative decay systems. Sigma3 might bind long dsRNAs
more strongly than P14; therefore, it protects hairpin tran-
scripts. P19, which does not bind long dsRNAs, fails to prevent
the generation of siRNAs from hairpin transcripts, instead it
inhibits silencing by sequestering hairpin-derived double-
stranded siRNAs.
Findings that P14 increases PVX and PoLV symptoms and
that virus-induced silencing is more intense in PoLV14-inoc-
ulated leaves than in PoLV-inoculated ones strongly indicate
that P14 acts as an efficient suppressor of virus-induced silenc-
ing. Viral siRNAs can be easily detected in PoLV-infected
leaves, suggesting that P14 inhibits virus-induced silencing by
targeting a step downstream of siRNA generation. As in vitro
P14 binds double-stranded siRNAs, we postulate that in virus-
infected cells P14, like P19, suppresses silencing by sequester-
ing double-stranded siRNAs. However, we failed to coimmu-
noprecipitate viral siRNAs from PoLV-infected leaves with
P14 polyclonal antibody (Z. Merai, unpublished data). It could
be due to technical difficulties (for instance, the P14 double-
stranded siRNA complex is weak and dissociates during ma-
nipulation or the P14 antibody is not suitable for coimmuno-
precipitation), and we cannot exclude that in vivo P14 does not
bind double-stranded siRNAs. An alternative model for P14-
mediated suppression of virus-induced silencing could be that
P14 binds the precursors of viral double-stranded siRNAs;
thus, it slightly delays the accumulation of siRNAs. Impor-
tantly, observation that viral siRNAs are relatively more abun-
dant in PoLV14-infected leaves than in PoLV-infected ones
is consistent with both suppression models.
Antiviral silencing operates as a cell-autonomous and sys-
temic response. Because viral RNAs accumulate to high lev-
els in PoLV14-transfected protoplasts (40) (Fig. 5B) even
though viral siRNAs are present (Fig. 5B), we conclude that
cell-autonomous antiviral silencing is unable to limit the accu-
mulation of the rapidly replicating virus. By contrast, infection
FIG. 6. P14 and P19 suppressors appear to be evolutionarily related. Multiple alignments of amino acid sequences deduced from many available
aureusvirus and tombusvirus ORF5 sequences. Defined gaps (see Materials S4 in the supplemental material) were incorporated into the deduced
protein sequences. PoLV (indicated as PLV), Cucumber leaf spot virus (CLSV), and Johnsongrass chlorotic stripe mosaic virus (JCSMV) are
aureusviruses, and other viruses included in the alignments are tombusviruses. CIRV refers to the P19 protein of CIRV, whereas CRSV shows
CymRSV P19. List of viruses, which were not used in this study but were included in the comparison are available in the supplemental material
(see Materials S4 and Table S1 in the supplemental material). Different colors show different groups of amino acids. Asterisks indicate amino acids
that are perfectly conserved in each aligned protein, while colons and periods refer to conservative and semiconservative substitutions, respectively.
Secondary structural elements defined for CIRV P19 are shown at the top.
7224 ME
of PoLV14 leads to a recovery phenotype (40) (Fig. 1A),
indicating that colonization of the host plant requires the sup-
pression of systemic silencing (1, 17, 18, 46, 49). We suggest
that, in wild-type infection, P14 inhibits the development of
systemic silencing by sequestering 21-nt double-stranded
siRNAs and/or by delaying the generation of siRNAs; thus,
PoLV spreads more quickly than the silencing signal and col-
onizes the plant. These models, in which P14 suppresses virus-
induced systemic silencing by binding double-stranded siRNAs
and/or precursor dsRNAs, predict that it inhibits systemic si-
lencing in a dose-dependent manner. Indeed, at a high tem-
perature (27°C) where siRNA generation is efficient (49), even
PoLV infection leads to a recovery phenotype, indicating that
at 27°C P14 fails to completely inhibit systemic silencing (Me-
rai, unpublished).
Evolution of dsRNA binding silencing suppressors of au-
reusviruses and tombusviruses. Both MP and suppressor pro-
teins of aureusviruses are homologous to the corresponding
tombusvirus proteins, indicating that the common ancestor of
these two genera already carried an ancestral silencing sup-
pressor nested in the MP. Because nested protein could not be
found in any related MP, we suggest that this ancestral sup-
pressor has evolved within the common ancestor of aureusvi-
rus-tombusvirus MP after it diverged from other 30-kDa MPs
but before the branching of Aureusvirus and Tombusvirus gen-
era. Moreover, as P19 and P14 suppressors are dsRNA binding
proteins and as the conserved suppressor regions are impor-
tant in forming the dsRNA binding structure of P19, it is likely
that the ancestral suppressor was a dsRNA binding protein.
P19 is a unique dsRNA binding protein because it binds
dsRNAs size selectively, while all other characterized dsRNA
binding proteins binds dsRNAs without strict size specificity.
Therefore, we speculate that the common ancestor of the au-
reusvirus-tombusvirus suppressor was a size-independent
dsRNA binding protein.
Suppressors have evolved to target antiviral responses, but
they have also been selected for causing as little damage as
possible to the host. It is proposed that, as size-specific double-
stranded siRNA binding is a result of such a dual selection, P19
can efficiently bind siRNAs that play a role in antiviral re-
sponse, while it might not interfere with long double-stranded
siRNA programmed silencing pathways such as RNA-medi-
ated epigenetic gene regulation (3, 33, 55). It is conceivable
that expression of a size-independent dsRNA binding protein
would cause additional damages, for instance, P14 might in-
terfere with RNA-directed epigenetic regulation or bind struc-
tured host mRNAs. Interestingly, the level of Sg2 RNA, from
which P14 is translated, declines after 2 to 3 d.p.i. in PoLV-
infected plants (40) (Fig. 5A), while CymRSV Sg2 is abundant
even at 10 d.p.i (48). It has been suggested that decreased
expression of PoLV Sg2 is a viral control measure to reduce
the toxicity of P14 (40). Indeed, infection with a PoLV mutant
that constitutively expresses Sg2 RNA results in rapid plant
death (40). It is appealing to speculate that the common an-
cestor suppressor might have evolved by two ways to reduce
the damage to the host. In tombusviruses, it has evolved into a
size-specific double-stranded siRNA binding suppressor, while
in PoLV, it could have evolved into a temporally/spatially con-
trolled suppressor.
Is dsRNA binding a frequent suppressor strategy? P14 sup-
pressed hairpin-induced silencing by preventing the accumula-
tion of hairpin transcript and siRNAs. Interestingly, transgenic
expression of peanut clump virus P15, potato virus X P25, and
turnip crinkle virus CP silencing suppressors in an Arabidopsis
line that expressed hairpin transcripts of chalcone synthase
also lead to similar inhibition of hairpin-induced silencing (14).
In all three cases, hairpin-derived siRNA levels were dramat-
ically reduced even though hairpin transcripts were not pro-
tected (14). These findings suggest that (some of) these
suppressors target silencing like P14. They might be size-inde-
pendent dsRNA binding proteins, which inhibit hairpin-in-
duced silencing by preventing DCL-mediated processing. If
these proteins bind dsRNAs like P14, they also form complexes
with double-stranded siRNAs; hence, they could suppress si-
lencing by sequestering double-stranded siRNAs. Indeed, P15
and CP suppressed siRNA-mediated silencing efficiently in
HeLa cells (14). Since these proteins are nonrelated (8), we
speculate that dsRNA binding is a frequent suppression strat-
egy, which has evolved independently many times. In this
study, we described a rapid mobility shift assay using crude
extracts prepared from either virus-infected or agroinfiltrated
leaves that is suitable for recognition and characterization of
dsRNA binding proteins. We think that this convenient
method could facilitate the identification of other dsRNA
binding viral suppressors.
We are grateful to M. Russo for kindly providing full-length cDNA
clones of PoLV and PoLV14 and to D. Baulcombe for the GFP plant
and 35SGFP constructs. We are especially grateful to J. Vargason and
T. Tanaka Hall for providing siRNAs and for help with conducting and
analyzing competition assays. We thank T. Tanaka Hall, G. Szittya,
and L. Lakatos for useful comments on the manuscript, G. Takacs for
help with figure preparations, and Edina Kapuszta for excellent tech-
nical assistance.
This research was supported by grants from the Hungarian Scientific
Research Fund (OTKA) (T15042787). D.S. was financed by the Bolyai
1. Al-Kaff, N. S., S. N. Covey, M. M. Kreike, A. M. Page, R. Pinder, and P. J.
Dale. 1998. Transcriptional and posttranscriptional plant gene silencing in
response to a pathogen. Science 279:2113–2115.
2. Baulcombe, D. C. 2004. RNA silencing in plants. Nature 431:356–363.
3. Baulcombe, D. C., and A. Molnar. 2004. Crystal structure of p19–a universal
suppressor of RNA silencing. Trends Biochem. Sci. 6:279–281.
4. Brigneti, G., O. Voinnet, W. X. Li, L. H. Ji, S. W. Ding, and D. C. Baulcombe.
1998. Viral pathogenicity determinants are suppressors of transgene silenc-
ing in Nicotiana benthamiana. EMBO J. 17:6739–6746.
5. Bucher, E., H. Hemmes, P. de Haan, R. Goldbach, and M. Prins. 2004. The
influenza A virus NS1 protein binds small interfering RNAs and suppresses
RNA silencing in plants. J. Gen. Virol. 85:983–991.
6. Carrington, J. C., and V. Ambros. 2003. Role of microRNAs in plant and
animal development. Science 301:336–338.
7. Chan, S. W., D. Zilberman, Z. Xie, L. K. Johansen, J. C. Carrington, and
S. E. Jacobsen. 2004. RNA silencing genes control de novo DNA methyl-
ation. Science 303:1336.
8. Chapman, E. J., A. I. Prokhnevsky, K. Gopinath, V. V. Dolja, and J. C.
Carrington. 2004. Viral RNA silencing suppressors inhibit the microRNA
pathway at an intermediate step. Genes Dev. 18:1179–1186.
9. Chen, J., W. X. Li, D. Xie, J. R. Peng, and S. W. Ding. 2004. Viral virulence
protein suppresses RNA silencing-mediated defense but upregulates the role
of micro RNA in host gene expression. Plant Cell 16:1302–1313.
10. Dalmay, T., L. Rubino, J. Burgyan, A. Kollar, and M. Russo. 1993. Func-
tional analysis of cymbidium ringspot virus genome. Virology 194:697–704.
11. Dalmay, T., R. Horsefield, T. H. Braunstein, and D. C. Baulcombe. 2001.
SDE3 encodes an RNA helicase required for post-transcriptional gene si-
lencing in Arabidopsis. EMBO J. 20:2069–2078.
12. Denli, A. M., and G. J. Hannon. 2003. RNAi: an ever-growing puzzle. Trends
Biochem. Sci. 28:196–201.
13. Ding, S. W., H. Li, R. Lu, F. Li, and W. X. Li. 2004. RNA silencing: a
conserved antiviral immunity of plants and animals. Virus Res. 102:109–115.
14. Dunoyer, P., C. H. Lecellier, E. A. Parizotto, C. Himber, and O. Voinnet.
2004. Probing the microRNA and small interfering RNA pathways with
virus-encoded suppressors of RNA silencing. Plant Cell 16:1235–1250.
15. Hamilton, A., O. Voinnet, L. Chappell, and D. C. Baulcombe. 2002. Two
classes of short interfering RNA in RNA silencing. EMBO J. 21:4671–4679.
16. Hamilton, A. J., and D. C. Baulcombe. 1999. A species of small antisense
RNA in posttranscriptional gene silencing in plants. Science 286:950–952.
17. Havelda, Z., C. Hornyik, A. Crescenzi, and J. Burgyan. 2003. In situ char-
acterization of Cymbidium Ringspot Tombusvirus infection-induced posttran-
scriptional gene silencing in Nicotiana benthamiana. J. Virol. 77:6082–6086.
18. Himber, C., P. Dunoyer, G. Moissiard, C. Ritzenthaler, and O. Voinnet.
2003. Transitivity-dependent and -independent cell-to-cell movement of
RNA silencing. EMBO J. 22:4523–4533.
19. Johansen, L. K., and J. C. Carrington. 2001. Silencing on the spot. Induction
and suppression of RNA silencing in the Agrobacterium-mediated transient
expression system. Plant Physiol. 126:930–938.
20. Kollar, A., and J. Burgyan. 1994. Evidence that ORF 1 and 2 are the only
virus-encoded replicase genes of cymbidium ringspot tombusvirus. Virology
21. Lakatos, L., G. Szittya, D. Silhavy, and J. Burgyan. 2004. Molecular mech-
anism of RNA silencing suppression mediated by p19 protein of tombusvi-
ruses. EMBO J. 23:876–884.
22. Lecellier, C. H., and O. Voinnet. 2004. RNA silencing: no mercy for viruses?
Immunol. Rev. 198:285–303.
23. Li, W. X., H. Li, R. Lu, F. Li, M. Dus, P. Atkinson, E. W. Brydon, K. L.
Johnson, A. Garcia-Sastre, L. A. Ball, P. Palese, and S. W. Ding. 2004.
Interferon antagonist proteins of influenza and vaccinia viruses are suppres-
sors of RNA silencing. Proc. Natl. Acad. Sci. USA 101:1350–1355.
24. Lichner, Z., D. Silhavy, and J. Burgyan. 2003. Double-stranded RNA-bind-
ing proteins could suppress RNA interference-mediated antiviral defences.
J. Gen. Virol. 84:975–980.
25. Lippmann, Z., and R. Martienssen. 2004. The role of RNA interference in
heterochromatic silencing. Nature 431:364–370.
26. Martelli, G. P., M. Russo, L. Rubino, and S. Sabanadzovic. 1995. Aureus-
virus, a novel genus in the family Tombusviridae. Arch. Virol. 143:1847–
27. Melcher, U. 2000. The ‘30K’ superfamily of viral movement proteins. J. Gen.
Virol. 81:257–266.
28. Miller, J. S., H. Damude, M. A. Robbins, R. D. Reade, and D. M. Rochon.
1997. Genome structure of cucumber leaf spot virus: sequence analysis
suggests it belongs to a distinct species within the Tombusviridae. Virus Res.
29. Mlotshwa, S., O. Voinnet, M. F. Mette, M. Matzke, H. Vaucheret, S. W. Ding,
G. Pruss, and V. B. Vance. 2002. RNA silencing and the mobile silencing
signal. Plant Cell 14:S289–301.
30. Mourrain, P., C. Beclin, T. Elmayan, F. Feuerbach, C. Godon, J. B. Morel,
D. Jouette, A. M. Lacombe, S. Nikic, N. Picault, K. Remoue, M. Sanial, T. A.
Vo, and H. Vaucheret. 2000. Arabidopsis SGS2 and SGS3 genes are required
for posttranscriptional gene silencing and natural virus resistance. Cell 101:
31. Palauqui, J. C., T. Elmayan, J. M. Pollien, H. Vaucheret. 1997. Systemic
acquired silencing: transgene-specific post-transcriptional silencing is trans-
mitted by grafting from silenced stocks to non-silenced scions. EMBO J.
32. Palauqui, J. C., and H. Vaucheret. 1998. Transgenes are dispensible for the
RNA degradation step of co-suppression. Proc. Natl. Acad. Sci. USA 95:
33. Papp, I., M. F. Mette, W. Aufsatz, L. Daxinger, S. E. Schauer, A. Ray, J. van
der Winden, M. Matzke, and A. J. Matzke. 2003. Evidence for nuclear
processing of plant micro RNA and short interfering RNA precursors. Plant
Physiol. 132:1382–1390.
34. Qiu, W., J. W. Park, and H. B. Scholthof. 2002. Tombusvirus P19-mediated
suppression of virus-induced gene silencing is controlled by genetic and
dosage features that influence pathogenicity. Mol. Plant Microbe Interact.
35. Qu, F., and T. J. Morris. 2002. Efficient infection of Nicotiana benthamiana
by Tomato bushy stunt virus is facilitated by the coat protein and maintained
by p19 through suppression of gene silencing. Mol. Plant Microbe Interact.
36. Qu, F., T. Ren, and T. J. Morris. 2003. The coat protein of turnip crinkle
virus suppresses posttranscriptional gene silencing at an early initiation step.
J. Virol. 77:511–522.
37. Ratcliff, F., B. D. Harrison, and D. C. Baulcombe. 1997. A similarity between
viral defense and gene silencing in plants. Science 276:1558–1560.
38. Reade, R., J. Miller, M. Robbins, Y. Xiang, and D. Rochon. 2003. Molecular
analysis of the cucumber leaf spot virus genome. Virus Res. 91:171–179.
39. Roth, B. M., G. J. Pruss, and V. B. Vance. 2004. Plant viral suppressors of
RNA silencing. Virus Res. 102:97–108.
40. Rubino, L., and M. Russo. 1997. Molecular analysis of the pothos latent virus
genome. J. Gen. Virol. 78:1219–1226.
41. Rubino, L., M. Russo, and G. P. Martelli. 1995. Sequence analysis of pothos
latent virus genomic RNA. J. Gen. Virol. 76:2835–2839.
42. Russo, M., J. Burgyan, and G. P. Martelli. 1994. Molecular biology of
tombusviridae. Adv. Virus Res. 44:381–428.
43. Ryabov, E. V., R. van Wezel, J. Walsh, and Y. Hong. 2004. Cell-to-cell, but
not long-distance, spread of RNA silencing that is induced in individual
epidermal cells. J. Virol. 78:3149–3154.
44. Schmidt, H. A., K. Strimmer, M. Vingron, and A. von Haeseler. 2002.
TREE-PUZZLE: maximum likelihood phylogenetic analysis using quartets
and parallel computing. Bioinformatics 18:502–504.
45. Silhavy, D., A. Molnar, A. Lucioli, G. Szittya, C. Hornyik, M. Tavazza, and
J. Burgyan. 2002. A viral protein suppresses RNA silencing and binds si-
lencing-generated, 21- to 25-nucleotide double-stranded RNAs. EMBO J.
46. Silhavy, D., and J. Burgyan. 2004. Effects and side-effects of viral RNA
silencing suppressors on short RNAs. Trends Plant Sci. 9:76–83.
47. Swofford, D. L. 2002. PAUP*. Phylogenetic analysis using parsimony (*and
other methods). Sinauer Associates, Sunderland, Mass.
48. Szittya, G., A. Molnar, D. Silhavy, C. Hornyik, and J. Burgyan. 2002. Short
defective interfering RNAs of tombusviruses are not targeted but trigger
post-transcriptional gene silencing against their helper virus. Plant Cell 14:
49. Szittya, G., D. Silhavy, A. Molnar, Z. Havelda, A. Lovas, L. Lakatos, Z.
Banfalvi, and J. Burgyan. 2003. Low temperature inhibits RNA silencing-
mediated defence by the control of siRNA generation. EMBO J. 22:633–640.
50. Takeda, A., K. Sugiyama, H. Nagano, M. Mori, M. Kaido, K. Mise, S. Tsuda,
and T. Okuno. 2002. Identification of a novel RNA silencing suppressor, NSs
protein of Tomato spotted wilt virus. FEBS Lett. 532:75–79.
51. Tariq, M., and J. Paszkowski. 2004. DNA and histone methylation in plants.
Trends Genet. 20:244–251.
52. Thomas, C. L., V. Leh, C. Lederer, and A. J. Maule. 2003. Turnip crinkle
virus coat protein mediates suppression of RNA silencing in Nicotiana
benthamiana. Virology 306:33–41.
53. Thompson, J. D., T. J. Gibson, F. Plewniak, F. Jeanmougin, and D. C.
Higgins. 1997. The CLUSTAL_X windows interface: flexible strategies for
multiple sequence alignment aided by quality analysis tools. Nucleic Acids
Res. 25:4876–4882.
54. Tijsterman, M., and R. H. Plasterk. 2004. Dicers at RISC; the mechanism of
RNAi. Cell 117:1–3.
55. Vargason, J. M., G. Szittya, J. Burgyan, and T. M. Tanaka Hall. 2003. Size
selective recognition of siRNA by an RNA silencing suppressor. Cell 115:
56. Voinnet, O., and D. C. Baulcombe. 1997. Systemic signalling in gene silenc-
ing. Nature 389:553.
57. Voinnet, O., Y. M. Pinto, and D. C. Baulcombe. 1999. Suppression of gene
silencing: a general strategy used by diverse DNA and RNA viruses of plants.
Proc. Natl. Acad. Sci. USA 96:14147–14152.
58. Voinnet, O., C. Lederer, and D. C. Baulcombe. 2000. A viral movement
protein prevents spread of the gene silencing signal in Nicotiana benthami-
ana. Cell 103:157–167.
59. Waterhouse, P. M., M. B. Wang, and T. Lough. 2001. Gene silencing as an
adaptive defence against viruses. Nature 411:834–842.
60. Xie, Z., K. D. Kasschau, and J. C. Carrington. 2003. Negative feedback
regulation of Dicer-Like1 in Arabidopsis by microRNA-guided mRNA deg-
radation. Curr. Biol. 13:784–789.
61. Xie, Z., L. K. Johansen, A. M. Gustafson, K. D. Kasschau, A. D. Lellis, D.
Zilberman, S. E. Jacobsen, and J. C. Carrington. 2004. Genetic and func-
tional diversification of small RNA pathways in plants. PLoS Biol. 2:E104.
62. Ye, K., L. Malinina, and D. J. Patel. 2003. Recognition of small interfering
RNA by a viral suppressor of RNA silencing. Nature 426:874–878.
63. Yue, Z., and A. J. Shatkin. 1997. Double-stranded RNA-dependent protein
kinase (PKR) is regulated by reovirus structural proteins. Virology 234:364–
64. Zhang, F., and A. E. Simon. 2003. Enhanced viral pathogenesis associated
with a virulent mutant virus or a virulent satellite RNA correlates with
reduced virion accumulation and abundance of free coat protein. Virology
7226 ME
    • "P19 is therefore a clear candidate for participation in a host-pathogen arms race [7, 9]. Moreover, the p19 gene is overprinted on the MP gene in all known tombusvirus genomes [27] (Fig 1A andFigure A in S1 File). The structure of p19 is inherently linked to its function—it is an obligate dimer with respect to siRNA size selection and binding [23, 24]—and p19 appears to exhibit greater sequence diversity than MP. "
    [Show abstract] [Hide abstract] ABSTRACT: Evolutionary arms races between pathogens and their hosts may be manifested as selection for rapid evolutionary change of key genes, and are sometimes detectable through sequence-level analyses. In the case of protein-coding genes, such analyses frequently predict that specific codons are under positive selection. However, detecting positive selection can be non-trivial, and false positive predictions are a common concern in such analyses. It is therefore helpful to place such predictions within a structural and functional context. Here, we focus on the p19 protein from tombusviruses. P19 is a homodimer that sequesters siRNAs, thereby preventing the host RNAi machinery from shutting down viral infection. Sequence analysis of the p19 gene is complicated by the fact that it is constrained at the sequence level by overprinting of a viral movement protein gene. Using homology modeling, in silico mutation and molecular dynamics simulations, we assess how non-synonymous changes to two residues involved in forming the dimer interface-one invariant, and one predicted to be under positive selection-impact molecular function. Interestingly, we find that both observed variation and potential variation (where a non-synonymous change to p19 would be synonymous for the overprinted movement protein) does not significantly impact protein structure or RNA binding. Consequently, while several methods identify residues at the dimer interface as being under positive selection, MD results suggest they are functionally indistinguishable from a site that is free to vary. Our analyses serve as a caveat to using sequence-level analyses in isolation to detect and assess positive selection, and emphasize the importance of also accounting for how non-synonymous changes impact structure and function.
    Full-text · Article · Feb 2016
    • "Moreover, P1 of RYMV also has ability to inhibit DCLs activities including both DCL4 and DCL3 (Lacombe et al., 2010; Weinheimer et al., 2010; Gillet et al., 2013). Some VSRs, such as Pothos latent aureusvirus (PolV) P14, TCV p38, CMV 2b, and RSV NS3 have been described to bind dsRNAs in a size-independent manner and therefore to block vsiRNA maturation (Merai et al., 2005; Deleris et al., 2006; Goto et al., 2007; Xiong et al., 2009). Furthermore, other VSRs such as RDV Pns10, CMV 2b, and Rice yellow stunt virus (RYSV) P6 block secondary siRNA biogenesis by downregulating RDR6 expression or suppressing the activity of RDR1, which are critical for secondary siRNAs synthesis (Diaz-Pendon et al., 2007; Ren et al., 2010; Guo et al., 2013). "
    [Show abstract] [Hide abstract] ABSTRACT: RNA silencing, an evolutionarily conserved and sequence-specific gene-inactivation system, has a pivotal role in antiviral defense in most eukaryotic organisms. In plants, a class of exogenous small RNAs (sRNAs) originating from the infecting virus called virus-derived small interfering RNAs (vsiRNAs) are predominantly responsible for RNA silencing-mediated antiviral immunity. Nowadays, the process of vsiRNA formation and the role of vsiRNAs in plant viral defense have been revealed through deep sequencing of sRNAs and diverse genetic analysis. The biogenesis of vsiRNAs is analogous to that of endogenous sRNAs, which require diverse essential components including dicer-like (DCL), argonaute (AGO), and RNA-dependent RNA polymerase (RDR) proteins. vsiRNAs trigger antiviral defense through post-transcriptional gene silencing (PTGS) or transcriptional gene silencing (TGS) of viral RNA, and they hijack the host RNA silencing system to target complementary host transcripts. Additionally, several applications that take advantage of the current knowledge of vsiRNAs research are being used, such as breeding antiviral plants through genetic engineering technology, reconstructing of viral genomes, and surveying viral ecology and populations. Here, we will provide an overview of vsiRNA pathways, with a primary focus on the advances in vsiRNA biogenesis and function, and discuss their potential applications as well as the future challenges in vsiRNAs research.
    Full-text · Article · Nov 2015
    • "To disrupt RNA silencing machinery, virus encoded RNA silencing suppressors target several effectors of RNAi silencing pathway, such as viral RNA recognition, dicing, RISC assembly, RNA targeting and amplifi- cation [4,5] e.g. P14 of Pothos latent aureusvirus and P38 of Turnip crinkle virus have been shown to inhibit the processing of dsRNA to siRNA [6,7]. In another case, the P19 protein of tombusviruses prevents RNA silencing by siRNA sequestration through binding ds siRNA with a high affinity [8]. "
    [Show abstract] [Hide abstract] ABSTRACT: Viral encoded RNA silencing suppressor proteins interfere with the host RNA silencing machinery, facilitating viral infection by evading host immunity. In plant hosts, the viral proteins have several basic science implications and biotechnology applications. However in silico identification of these proteins is limited by their high sequence diversity. In this study we developed supervised learning based classification models for plant viral RNA silencing suppressor proteins in plant viruses. We developed four classifiers based on supervised learning algorithms: J48, Random Forest, LibSVM and Naïve Bayes algorithms, with enriched model learning by correlation based feature selection. Structural and physicochemical features calculated for experimentally verified primary protein sequences were used to train the classifiers. The training features include amino acid composition; auto correlation coefficients; composition, transition, and distribution of various physicochemical properties; and pseudo amino acid composition. Performance analysis of predictive models based on 10 fold cross-validation and independent data testing revealed that the Random Forest based model was the best and achieved 86.11% overall accuracy and 86.22% balanced accuracy with a remarkably high area under the Receivers Operating Characteristic curve of 0.95 to predict viral RNA silencing suppressor proteins. The prediction models for plant viral RNA silencing suppressors can potentially aid identification of novel viral RNA silencing suppressors, which will provide valuable insights into the mechanism of RNA silencing and could be further explored as potential targets for designing novel antiviral therapeutics. Also, the key subset of identified optimal features may help in determining compositional patterns in the viral proteins which are important determinants for RNA silencing suppressor activities. The best prediction model developed in the study is available as a freely accessible web server pVsupPred at
    Full-text · Article · May 2014
Show more