Content uploaded by Olivier Voinnet
Author content
All content in this area was uploaded by Olivier Voinnet on Aug 15, 2014
Content may be subject to copyright.
Organisms must constantly balance resources between
investment in growth and development on the one
hand and defence against biotic and abiotic stress on
the other. In eukaryotes, and in plants in particular,
RNA silencing plays a major part in this balance by
dynamically linking developmental programmes and
environmental responses to gene expression changes
through transcriptional gene silencing (TGS) and
post-transcriptional gene silencing (PTGS)1 (BOX1). In
plants, the targets of PTGS range from stress-responsive
genes to genes required for cell-type specification and
organ patterning2,3, whereas TGS functions primarily
to epigenetically repress the transcription of transpos-
able elements and their genetic remnants, a mechanism
that might also modulate the expression of transposable
element-proximal genes in response to environmental
stimuli or stress4.
The origins of RNA silencing are that of an ancient
mechanism that directly defends host cells against
foreign nucleic acids, including viruses and active
transposable elements5. This defence is stimulated
by double-stranded RNA (dsRNA), a signature mol-
ecule derived from amplification of invasive nucleic
acids, which is processed by the host into small RNAs
(sRNAs) that are 20–24 nucleotides (nt) in size. These
sRNAs are then used to guide the silencing of the viral
or transposable element RNA or DNA through PTGS
or TGS, respectively. RNA silencing has subsequently
expanded and diversified to encompass the regulation
of endogenous genes through various sources of host-
encoded dsRNA6 (BOX1), placing this process at a cross-
road between defence, development and adaptation
tostress.
Independent studies of plant immunity to non-viral
pathogens have revealed how sophisticated signalling
networks underpin a generic basal defence mechanism
that is activated on perception of many types of path-
ogen-associated molecular patterns (PAMPs)7. This basal
defence layer, called PAMP-triggered immunity (PTI),
accounts for a co-evolutionary arms race that has shaped
pathogen genomes to produce virulence factors, called
effectors, which suppress PTI8 (BOX2). In turn, plants
respond to PTI suppression using a highly specific
and potent form of immunity called effector-triggered
immunity (ETI), in which dedicated plant resistance
(R) proteins recognize specific effector alleles within
microbial populations9 (BOX2). Because they represent
protein-based rather than RNA-based defence mecha-
nisms, historically PTI and ETI have been considered
to be largely independent from antiviral RNA silencing.
A recent body of evidence indicates, however, that this
view requires considerable amendment, because PTGS
Swiss Federal Institute of
Technology Zurich
(ETH‑Zurich), Department of
Biology, Zurich, Switzerland.
Correspondence to O.V.
e‑mail: voinneto@ethz.ch.
doi:10.1038/nrmicro3120
Pathogen-associated
molecular patterns
(PAMPs; alternatively,
microorganism-associated
molecular patterns). Conserved
and essential pathogen
signature molecules that
include flagellin, lipopoly-
saccharide, elongation factors
and double-stranded RNA.
RNA silencing suppression by plant
pathogens: defence, counter-defence
and counter-counter-defence
Nathan Pumplin and Olivier Voinnet
Abstract | RNA silencing is a central regulator of gene expression in most eukaryotes and acts
both at the transcriptional level through DNA methylation and at the post-transcriptional
level through direct mRNA interference mediated by small RNAs. In plants and invertebrates,
the same pathways also function directly in host defence against viruses by targeting viral
RNA for degradation. Successful viruses have consequently evolved diverse mechanisms to
avoid silencing, most notably through the expression of viral suppressors of RNA silencing.
RNA silencing suppressors have also been recently identified in plant pathogenic bacteria
and oomycetes, suggesting that disruption of host silencing is a general virulence strategy
across several kingdoms of plant pathogens. There is also increasing evidence that plants
have evolved specific defences against RNA-silencing suppression by pathogens, providing
yet another illustration of the never-ending molecular arms race between plant pathogens
and their hosts.
REVIEWS
NATURE REVIEWS
|
MICROBIOLOGY VOLUME 11
|
NOVEMBER 2013
|
745
FOCUS ON PLANT–MICROBE INTERACTIONS
© 2013 Macmillan Publishers Limited. All rights reserved
AGO4
Nature Reviews | Microbiology
MIRNA
miRNA
precursor
Long
dsRNA
Long
dsRNA
Aberrant
RNA or TAS
precursor
Mature
miRNA
HEN1
21- or 22-nt
siRNA
HEN1
24-nt
rasiRNA
DRM1/2
a b c
RDR2
TE repeat
Pol IV
Pol II
Me
Me
Me Me
Me Me
Me
AAA
AGO1
AAA
AGO1
Pol V
RDR6
SGS3
DCL4DCL3 DCL2
DRB4
HEN1
Me
HEN1
Me HEN1
Me
HEN1
DCL1
HYL1
Oomycetes
Eukaryotic fungal-like
organisms of the kingdom
Stramenopila that include
the causal agent of the Irish
potato famine, Phytophthora
infestans.
and TGS mediated by endogenous sRNAs are now
emerging as key regulators of PTI and ETI signalling
and R gene expression10–12. Moreover, as was originally
discovered with most plant viruses, several classes of
microorganisms, including bacteria and oomycetes, were
recently found to produce suppressors of RNA silencing
as part of their arsenal of virulence effectors13,14. Here,
we review the advances made in our understanding of
antiviral RNA silencing as a paradigm of induction and
suppression of PTGS. We further expand and contrast
these findings to those made with other classes of patho-
gens and show how RNA silencing and classic PTI and
Box 1 | Endogenous RNA silencing pathways in Arabidopsis thaliana
RNA silencing is ubiquitously triggered by double-stranded RNA (dsRNA). This molecule can arise endogenously from
transcription of imperfectly matched hairpins defining the microRNA (miRNA) genes, from self-complementary RNA
produced by inverted-repeat (IR) loci or by overlapping, bidirectional or convergent transcription1. Aberrant RNA, notably
produced by transposons and their remnants scattered in the genome, can also be amplified into dsRNA by the action of
one of six Arabidopsis thaliana-encoded RNA-dependent RNA polymerases (RDRs)142. The dsRNA is then processed by
one of four ribonuclease III-type Dicer-like (DCL) enzymes into 21–24-nucleotide (nt) RNA duplexes that fall into two
distinct classes20,125,143: small interfering RNAs (siRNAs) form populations that arise from sequential dicing of perfect, or
near perfect, long dsRNA, whereas miRNAs are excised as discrete species from their imperfectly folded precursor RNAs
and can individually accumulate to extremely high levels within plant tissues. All known siRNA and miRNA duplexes are
protected from degradation by 2′-O‑methylation mediated by the methyltransferase HUA ENHANCER1 (HEN1)144,145.
Upon processing by DCLs (‘dicing’), stabilized siRNA or miRNA duplexes are incorporated into one of ten Argonaute
(AGO) proteins to form an RNA-induced silencing complex (RISC)32. RISCs can target siRNA- or miRNA-complementary
mRNAs and induce their post-transcriptional gene silencing (PTGS) by endonucleolytic cleavage (‘slicing’) or translational
repression; RISCs can also induce transcriptional gene silencing (TGS) by DNA methylation and chromatin modifications
of target loci36,146.
In A.thaliana, members of the DCL and AGO families display some specialization: most miRNAs are processed in the
nucleus by DCL1 and its dsRNA-binding cofactor protein, HYPONASTIC LEAVES 1 (HYL1)147 (see the figure, part a) into
21-nt or, less frequently, 22-nt duplexes, consisting of a guide strand and a labile passenger strand, known as miRNA*,
which is usually degraded. On their transport to the cytoplasm, mature miRNAs are loaded primarily into AGO1 to
regulate, by PTGS, expression of target mRNAs involved in many aspects of plant biology, including development and
adaptation to stress2,3. Long, perfectly complementary dsRNA produced by RDR2 derives mostly from transcripts
produced by the DNA-dependent RNA polymerase IV (Pol IV) at DNA repeats and transposon loci, and is processed by
DCL3 into 24-nt-long repeat-associated small RNAs (rasiRNAs)146 (see the figure, part b). rasiRNAs are loaded primarily
into AGO4, and secondarily into AGO6 or AGO9 to initiate RNA-directed DNA methylation (RdDM) through DOMAINS
REARRANGED METHYLTRANSFERASE (DRM) proteins at the rasiRNA-generating loci, using transcripts produced by the
DNA-dependent Pol V as scaffolds148–151. These loci can consequently undergo TGS, which might also affect expression of
neighbouring protein-encoding genes.
Other endogenous siRNAs mediate PTGS on their loading into AGO1 or AGO2. These siRNAs are processed from
dsRNA derived from non-coding RNA, mRNA or natural antisense RNA through the hierarchical and redundant action of
DCL4 and the related DCL2, producing, respectively, 21- and 22-nt siRNAs5 (see the figure, part c). Trans-acting siRNA
(tasiRNA) biogenesis represents an interesting case whereby single-stranded non-coding tasiRNA precursors first
undergo slicing guided by an AGO-loaded miRNA, usually 22 nt in size152–154. These features stimulate the recruitment, at
the slicing site, of RDR6, which copies the cleaved RNA fragment into long dsRNA. Subsequent processing by DCL4 leads
to a population of 21-nt tasiRNAs targeting complementary transcripts that often belong to multi-gene families. DRB4,
DOUBLE-STRANDED-RNA-BINDING PROTEIN 4; SGS3, SUPPRESSOR OF GENE SILENCING 3; TE, transposon element.
REVIEWS
746
|
NOVEMBER 2013
|
VOLUME 11 www.nature.com/reviews/micro
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved
Nature Reviews | Microbiology
HR threshold
Resistance
threshold
Low
High PTI
R R
ETI ETI
Effector 1
Defence amplitude
Effector 2
PAMP
HR threshold
Resistance
threshold
Low
High PTI
R R
ETI ETI
Effector 1
Defence amplitude
Effector 2
PAMP
Viral suppressors of RNA
silencing
(VSRs). Proteins encoded
by viruses that function as
virulence factors by inhibiting
post-transcriptional gene
silencing pathways in plants
and invertebrates.
ETI pathways have become increasingly intermingled in
the co-evolutionary arms race between plants and their
pathogens.
Plant RNA viruses and PTGS
Most plant viruses have RNA genomes that contain
imperfect regulatory stem-loops and are copied into
complementary dsRNA replication intermediates (RIs)
by virus-encoded RNA-dependent RNA polymerases
(RDRs). As an intrinsic feature of virus genome expres-
sion and replication, this dsRNA can be designated a
virus-associated molecular pattern (VAMP; a form of
PAMP). VAMPs are generically recognized by Dicer-
like (DCL) enzymes, which then produce virus-derived
small interfering RNAs (vsiRNAs) that, upon load-
ing into Argonaute (AGO) proteins, promote antiviral
defence through RNA silencing15 (BOX1).
This innate immune response is remarkably versa-
tile because, being solely programmed by structural and
nucleotide-sequence genomic features, it can respond
to virtually any plant virus. In fact, so widespread and
potent is this defence response, it has driven most, if
not all, plant viruses to evolve viral suppressors of RNA
silencing (VSRs) that attenuate or completely inhibit
this process. Despite its universality, antiviral silenc-
ing remains mechanistically incompletely understood,
as our knowledge derives mostly from reverse genetic
studies that were conducted primarily in the model plant
Arabidopsis thaliana infected with viruses with genomes
that had probably been shaped by divergent plant hosts,
despite infecting A.thaliana in the wild16. Furthermore,
these studies have relied heavily on only a handful of
viruses for which VSR-deficient strains are available,
a necessity to fully reveal the antiviral effects of RNA
silencing17.
A.thaliana, in which the functions of DCL4 or its
cofactor DOUBLE-STRANDED-RNA-BINDING
PROTEIN 4 (DRB4) are compromised, are hyper-
susceptible to infection with multiple RNA viruses,
a phenotype that is strongly exacerbated in dcl2 dcl4
double mutants, although single dcl2 mutants display
little effect on their own18–22. By contrast, virus accu-
mulation is unchanged in mutants in DCL1 or DCL3,
which are involved in microRNA (miRNA) and repeat-
associated small interfering RNA (rasiRNA) biogenesis,
respectively (BOX1). Hypersusceptibility to infection
with RNA viruses is also observed in HUA enhancer 1
(hen1)23, ago1 (REF.24) and, as discovered more recently,
ago2 single mutants, and is enhanced in ago1 ago2 dou-
ble mutants22,25–28. Hypersusceptibility to infection with
VSR-deficient Cucumber mosaic virus (CMV) was not
observed, however, in any of the remaining ago mutants
or combinations thereof. Consistent with these genetic
data, 2′-O-methylated vsiRNAs co-immunoprecipitate
with AGO1 and AGO2, but not, for instance, with
AGO4, AGO6 or AGO9, the cognate rasiRNA effec-
tors of RNA-directed DNA methylation and TGS22,29,30
(BOX1). Thus, a model for antiviral defence emerges in
which viral dsRNA is processed primarily by DCL4
assisted by DRB4, and secondarily by DCL2. Methylated
and protected from degradation by HEN1 (BOX1), both
classes of vsiRNAs (primary vsiRNAs, which are pro-
duced in the absence of a host RDR, and secondary
vsiRNAs, which are produced by a host RDR), are then
incorporated into AGO1 and AGO2 to target comple-
mentary viral RNA through PTGS, although the exact
mechanisms involved are still unclear (FIG.1).
Mechanisms and localization of antiviral silencing. In
principle, ‘dicing’ of dsRNA viral RIs (that is, process-
ing by DCLs) could be sufficient to mediate antiviral
silencing. However, ago1 ago2 double mutants, despite
Box 2 | R gene-mediated immunity
The plant immune system comprises two levels of microbial recognition (also
summarized in FIG.6). First, the presence of extracellular pathogen-associated
molecular patterns (PAMPs) or microorganism-associated molecular patterns
(MAMPs) derived from bacteria, fungi or oomycetes is detected by a class of plasma
membrane-localized pattern-recognition receptors (PRRs) that activate a basal
resistance termed PAMP-triggered immunity (PTI)9 (see the figure). The archetypal PTI
model is initiated by the transmembrane receptor FLAGELLIN-SENSITIVE 2 (FLS2)
following recognition of the conserved bacterial flagellin-derived peptide, flg22,
resulting in a mitogen-activated protein kinase activation cascade, oxidative burst,
callose deposition and the induction of basal defence genes7,155,156. The existence of PTI
has instigated the necessity for plant pathogens to evolve a suite of diverse effector
molecules that are secreted into host cells by, for example, the bacterial typeIII
secretion system (T3SS), to interfere with PTI. In most cases, these counter-defence
effectors are virulence factors because they promote microbial growth and disease;
accordingly, pathogens that are unable to deliver effectors into the host cell, such as
T3SS-deficient bacteria, cannot thrive on their hosts and are deemed ‘non-virulent’9.
As a counter-counter-defence to the action of microbial effectors, plants have
developed a large family of structurally conserved nucleotide binding site leucine-rich-
repeat resistance (R) proteins that recognize pathogen effectors by monitoring (or
‘guarding’) the integrity of their endogenous targets (often referred to as ‘guardees’
or ‘decoys’)157. Activation of R proteins by cognate microbial effectors triggers a
strong defence response called effector-triggered immunity (ETI). Being prone to
recombination, R genes constantly evolve within plant populations, such that specific
Ralleles become selected to confer optimal recognition of dominant microbial
effectors found in particular ecological niches158. This selection pressure, in turn, drives
microorganisms to evolve mutations in effector genes in order to minimize or prevent
the detection of their products by R proteins. Effectors recognized by plant R proteins
are called avirulence factors because the pathogens secreting these factors (called
avirulent pathogens) usually fail to colonize their hosts owing to strong ETI activation.
ETI typically includes the accumulation and/or secretion of pathogenesis-related
proteins and antimicrobial compounds, often coinciding with a form of localized
necrosis, or programmed cell death, termed the hypersensitive response, which is
thought to restrict biotrophic pathogen growth. ETI also involves salicylic acid, ethylene
and nitric oxide signalling, as well as systemic acquired resistance, which enhances
resistance against a wide range of pathogens in tissues distal to the primary infection
site. Although R gene-based defence mechanisms are highly effective, as they
recognize effectors from many types of pathogens, their uncontrolled activation
comes at considerable fitness costs because hyperactive or constitutive ETI causes
stunting or even lethality in plants. Consequently, R gene expression and activity is
tightly regulated by plants
at multiple levels, which
include, as discovered
recently, RNA silencing at
both the transcriptional
gene silencing and post-
transcriptional gene
silencing levels159,160.
HR, hypersensitive response.
REVIEWS
NATURE REVIEWS
|
MICROBIOLOGY VOLUME 11
|
NOVEMBER 2013
|
747
FOCUS ON PLANT–MICROBE INTERACTIONS
© 2013 Macmillan Publishers Limited. All rights reserved
RDR1
RDR6
AGO1 ?
AGO2
Nature Reviews | Microbiology
AAA
DCL2
DCL4
DRB4
DCL2
DCL4
DRB4
a d
b
e
f
c
Mitochondrion
vsiRNA
Long dsRNA
Viral
mRNA
RI
Abberant viral
transcription byproducts
Translocation between cells
through plasmodesmata?
Rep
DCL4
DCL2
DRB4
AGO1
AGO2
AAA
AGO1
SGS3
Nucleus
Rough ER
accumulating abundant vsiRNAs owing to DCL action,
are hypersusceptible to infection with VSR-deficient
CMV, indicating that AGO functions are essential for
antiviral silencing22. Until recently, the importance of
‘slicing’ (that is, cleavage by AGO proteins) as opposed
to alternative modes of antiviral AGO action, such as
translational repression (BOX1; FIG. 1), had not been
addressed. Expression of a catalytic-null AGO2 allele
in A.thaliana ago2 knockout mutants was insufficient
to block accumulation of VSR-deficient Tur nip m osa ic
virus (TuMV), suggesting that vsiRNA-loaded AGO2
acts mainly through RNA slicing31 (FIG.1). The mode
of AGO1 action in antiviral defence remains compara-
tively less clear owing to the pleiotropic phenotypes of
ago1 mutants, which are largely due to the required role
for AGO1 in the miRNA pathway, which is not rescued
by slicer-deficient AGO1 alleles (BOX1). The synergistic
phenotype of ago1 ago2 double mutants indicates that
these AGOs have non-overlapping functions in anti-
viral defence, consistent with their phylogenetic dis-
tance32. Although AGO2 might be the primary operator
of vsiRNA action, AGO1 might act as a direct vsiRNA
effector and also influence other host defence pathways
that activate on pathogen-induced perturbation of the
endogenous miRNA pathway, as discussed in the final
section of thisReview.
The precise nature of the viral RNAs that are targeted
by AGO1 and AGO2 also remains unclear because plant
RNA viruses replicate within membrane invaginations
that not only protect their genomes from silencing but
also vary extensively, ranging from outer mitochondrial
membranes and outer plastid membranesto endoplas-
mic reticulum aggregates33. Direct targeting of replicat-
ing viral genomes would thus entail the existence of
active mechanisms recruiting AGOs to these various
membrane structures (FIG.1). Although such mecha-
nisms remain unknown, emerging evidence indicates
that AGO1 functions partially in association with plant
membranes: mutations affecting isoprenoid synthesis
and certain ago1 hypomorphic alleles reduce this mem-
brane association and impair AGO1 action34, and AGO1
was recently shown to translationally repress miRNA
targets in association with the rough endoplasmic retic-
ulum35. Alternatively to targeting replicating genomes,
antiviral AGOs could target viral transcripts during
their translation by cleavage and/or translational repres-
sion, effectively depriving viruses of proteins required
for genome amplification and coating36–39 (FIG.1). The
reported nuclear localization of the four A.thaliana
DCL proteins, including the antiviral DCL2 and DCL4
(REFS 40,41), also conflicts with the exclusive cytoplas-
mic localization of RNA viruses (FIG.1). This discrepancy
could be explained if some viral dsRNA is transferred
to the nucleus for processing or, alternatively, if antivi-
ral DCLs re-localize to the cytoplasm during infection
(FIG.1), as was shown for the nuclear DRB4, which forms
cytoplasmic punctae following infection with Turnip
yellow mosaic virus19.
In many virus infections, most of the vsiRNAs are
‘secondary’ vsiRNAs, which are processed by DCL4 and
DCL2 from dsRNA synthesized by host-encoded RDRs,
in contrast to the ‘primary’ vsiRNAs that are produced
from virus-encoded polymerase RNA templates. RDR6
and its associated coiled-coil domain-containing cofac-
tor, SUPPRESSOR OF GENE SILENCING 3 (SGS3),
Figure 1 | Molecular and cellular bases for antiviral silencing. Reverse genetic data
have unequivocally established the antiviral nature of RNA silencing through viral
double-stranded RNA (dsRNA) processing by dicer-like 4 (DCL4) and DCL2 and targeting
of viral genomes or transcripts by virus-derived small interfering (vsiRNA)-loaded
ARGONAUTE 1 (AGO1) and AGO2. However, the molecular and cellular mechanisms
underlying these two steps remain obscure. The figure depicts non-mutually exclusive,
alternative possibilities that are consistent with the current data. a–c | Possible modes
of DCL action. Dicing might occur directly on viral dsRNA formed during replication
(parta), linked to RNA-dependent RNA polymerase (RDR) activity in the cytoplasm (the
habitat of RNA viruses) (part b) or might occur on viral RNA entering the nucleus from
the cytoplasm before or after RDR action (part c). d–f | Possible modes of AGO action.
Viral RNA slicing by AGO2 might occur at the site of replication (here, mitochondrial
membranes), although whether AGO1 slices viral RNA remains unknown (indicated by
the question mark) (part d); viral RNA slicing might occur on viral transcripts that are
destined for translation (part e). Translational repression of viral transcripts on the
rough endoplasmic reticulum (ER) might occur through the action of AGO1 (part f).
Additionally, vsiRNAs might move from cell to cell through the plamodesmata to
immunize surrounding, naive tissues. DRB4, DOUBLE-STRANDED-RNA-BINDING
PROTEIN 4; Rep, replication-associated protein; RI, replication intermediate.
REVIEWS
748
|
NOVEMBER 2013
|
VOLUME 11 www.nature.com/reviews/micro
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved
MicroRNA
(miRNA). Discrete 21- or
22-nucleotide small RNAs,
arising from genome-encoded
imperfect foldback structures
processed by DCL1, which
negatively regulate
endogenous target mRNAs by
complementary base-pairing.
Repeat-associated small
interfering RNA
(rasiRNA). 24-nucleotide small
interfering RNAs generated
from genomic repeats and
transposon loci by DCL3, which
target DNA methylation and
chromatin modification at their
loci of origin.
Trans-acting siRNA
(tasiRNA). 21-nucleotide small
interfering RNAs produced by
sequential processing by DCL4
on long, perfectly
complementary
double-stranded RNA.
are key contributors to this amplification process18,42,43
(FIG.1). By contrast, both AGO1 and AGO2 are dispen-
sable for amplification, at least during infection by VSR-
deficient CMV, indicating that slicing is not required to
generate the viral RNAs that are used by RDR6 and SGS3
as templates22. The biogenesis of secondary vsiRNAs
probably differs, therefore, from the pathway that pro-
duces trans-acting siRNAs (tasiRNAs) (BOX1); however, it is
difficult to establish whether the biogenesis of secondary
vsiRNAs requires primary vsiRNAs because both share
the same biogenesis machinery. A hint to the nature of
such templates comes from studies of transgene-induced
silencing, which suggest that RDR6 activity is stimulated
by an excess of aberrant RNAs lacking a poly(A) tail or a
cap — features that are likely to be displayed by aborted
viral transcription and/or replication products44–46. The
additional requirement for the salicylate-induced RDR1
and, to some extent, for RDR2, emphasizes the impor-
tance of denovo dsRNA synthesis and amplification in
antiviral silencing21,47–50. As was demonstrated for their
transgenic or endogenous counterparts51,52, amplified
vsiRNAs are believed to move between cells and over
long distances through the phloem to immunize distant
tissues ahead of viralspread.
A final unanswered yet central question is whether
all vsiRNAs are equally efficient in mediating antivi-
ral silencing. Recent genetic studies indeed suggest
that secondary DCL2-dependent 22-nt siRNAs have
inefficient viral targeting activities and might be more
specialized, perhaps, in mediating systemic silencing
effects21,22. Parallel vsiRNA deep-sequencing and high-
throughput viral RNA degradome analyses (a method
to map sliced RNA fragments) also suggest that only a
small, and perhaps select, fraction of vsiRNAs accounts
for cleavage of viral RNA38. Further complicating the
picture, the size and 5ʹ-terminal nucleotide identity of
vsiRNAs strongly influence their loading into specific
AGOs, with 21- and 22-nt vsiRNAs with a 5ʹ U and 5ʹ A
being primarily incorporated into antiviral AGO1 and
AGO2, respectively; 24-nt vsiRNAs with a 5ʹ U and 21-nt
vsiRNAs with a 5ʹ C, by contrast, would probably load into
AGO4 and AGO5, respectively, which are not involved
in antiviral defence53,54 (BOX1). It is, in fact, conceivable
that positive selection might be exerted on viral genomes
to bias abundantly diced vsiRNAs towards 5ʹ nucleotide
identities that avoid loading into antiviralAGOs.
Suppression and evasion of antiviral silencing. An
invariable tenet of the never-ending molecular arms
race between pathogens and their hosts is the ability of
the pathogens to avoid, actively suppress or even hijack
host defence pathways. Antiviral silencing is certainly
no exception, as was originally established when the
potyviral helper-component proteinase (HcPro) and
cucumoviral 2b protein, which had both previously been
identified as virulence factors, were found to suppress
transgene silencing55–57. The discovery that RNA silenc-
ing protects plants against infection by most viruses
coincided with the discovery that many additional viral
pathogenicity determinants act as VSRs58. VSRs have
now been isolated from nearly all phytovirus families
and arose independently multiple times to target RNA
silencing at different points and through diverse mecha-
nisms: an example of convergent evolution. The study
of VSRs not only greatly improves our understanding of
plant–virus interactions but also illuminates fundamen-
tal aspects of the host silencing machinery itself. These
notions are illustrated below and in FIG.2 with selected
VSRs for which the molecular mode of action has now
been — at least partly — elucidated.
A first class of VSRs inactivates, competes with or
compromises the integrity of silencing effector proteins.
For instance, the Turnip crinkle virus (TCV)-encoded
P38 protein forms homodimers that bind AGO1 and
possibly AGO2, but not AGOs that are unlinked to
antiviral defence25,29 (FIG.2a). The P38–AGO1 interac-
tion depends on Gly-Trp (GW) ‘AGO-hook’ residues
and compromises AGO1 loading with siRNAs, but
not with miRNAs, in P38 transgenic plants59. AGO1 is
ubiquitylated by the polerovirus-encoded F-box protein
P0 through the host SPHASE KINASE-ASSOCIATED
PROTEIN1 (SKP1)–CULLIN1 complex60,61 (FIG.2b).
The ensuing AGO1 degradation is insensitive to
inhibitors of the ubiquitin–proteasome system but
affected by drug treatments or mutations impairing
autophagy62. In fact, P0 hijacks a normal physiological
process whereby unloaded AGO1 undergoes selective
autophagy in healthy plants, a regulatory process that
also affects AGO and GW-rich proteins in human cells
and Caenorhabditis elegans63,64, emphasizing the value
of VSRs as molecular probes of RNA silencing mecha-
nisms in general. The P25 protein of Potato virusX
(PVX) also promotes degradation of RNA silencing
effectors, including the antiviral AGO1 and AGO2, but
in a manner sensitive to treatments with the protea-
some inhibitor MG132 (REF.65) (FIG. 2b). CMV 2b also
displays AGO-binding capacity; however, its VSR func-
tion depends rather on its RNA-binding activity66. VSRs
can also impair the production of siRNAs, as illustrated
by the transactivator protein (TAV; also known as P6) of
Cauliflower mosaic virus (CaMV, a DNA virus), which
interacts directly with the DCL4 cofactor DRB4 (REF.67)
(FIG.2c). VSRs acting specifically at the level of silencing
amplification possibly include the Rice yellow stunt virus
(RYSV) protein P6, which suppresses systemic, but not
intracellular, transgene silencing by directly interacting
with RDR6 and blocking secondary siRNA synthesis68.
The V2 protein of the DNA virus Tomato yellow leaf
curl virus (TYLCV) also interacts functionally with the
tomato homologue of the A.thaliana RDR6 cofactor
SGS3 to suppress amplified RNA silencing69 (FIG. 2d).
VSR action at the level of sRNA stability is illustrated
by the Tobacco mosaic virus (TMV)-encoded 126 kDa
replicase subunit P126, which inhibits HEN1-mediated
stabilization of vsiRNAs and siRNAs23 (FIG.2e; BOX1), an
effect that is also promoted by TCV infection. Direct
interference with vsiRNA stability was reported in
the dsRNA-specific class1 RNA endoribonuclease III
(RNase III) of Sweet potato chlorotic stunt virus, which
cleaves synthetic siRNA duplexes of 21-, 22- and 24-bp
invitro into approximately 14-bp products that are
inactive in mediating silencing70 (FIG.2f).
REVIEWS
NATURE REVIEWS
|
MICROBIOLOGY VOLUME 11
|
NOVEMBER 2013
|
749
FOCUS ON PLANT–MICROBE INTERACTIONS
© 2013 Macmillan Publishers Limited. All rights reserved
RDR6
AGO1AGO1AGO1
AGO1AGO1
AGO1
AGO1
Nature Reviews | Microbiology
a
c
e
gh
d
b
P0
P25 Ub
Ub
Ub
AGO1
mRNA
Most miRNAs
and vsiRNAs
sequestered
siRNA
production
Proteasome
Autophagosome Lysosome
NLS AAA
TAV
Nucleus
TAVDRB4
TAV
V2 SGS3
f
RNase III
21 nt
14 nt
21 nt
MIR168
GW
GW
P38 P38
GA
GA
P38 P38
HEN1 P126
UUU UUU
P19
PSTVd
DCL DCL DCL
AGO
AAA
AGO1
One of the earliest and most insightful depictions
of VSR action came as a result of the crystallization of
the tombusvirus P19 protein bound to a 21-bp siRNA
substrate71,72. P19 forms head-to-tail homodimers in
which the distance between key Trp residues provides
the basis for a ‘molecular caliper’ that binds to vsiR-
NAs and siRNAs, as well as endogenous miRNAs,
with high affinity, preventing their loading into AGO1
(FIG.2g). Remarkably, miR168, unlike other endogenous
miRNAs, is not efficiently bound by P19, resulting in the
selective increased loading of miR168 into AGO1 and
the related AGO10 (REF.73). Because miR168 directly
downregulates AGO1 mRNA stability and translation
as part of a regulatory feedback loop74,75, the net effect
of this selective binding process is a sharp reduction in
cellular AGO1 levels in addition to the direct siRNA
sequestration by P19 (FIG.2g). Tombusvirus infection also
specifically induces MIR168 transcription in a silencing
suppression-dependent manner, thereby further increas-
ing the levels of miR168 available for AGO1 downregu-
lation73. Enhanced MIR168 transcription and lowered
AGO1 levels are also observed during infections by
tombusvirus-unrelated viruses and are recapitulated by
transient expression of their cognate VSRs, supporting
the emerging idea that diverse VSRs convergently hijack
endogenous silencing to suppress the antiviral silencing
pathway76 (BOX3).
RNA-based as opposed to protein-based strategies
can be also deployed by viruses to evade rather than
suppress silencing. For instance, Potato spindle tuber
viroid (PSTVd), a subviral pathogen with a non-coding,
complementary circular RNA genome, acquires an
extended quasi-rod-shaped structure that is an effec-
tive DCL substrate but is largely inaccessible to AGO
action77–80 (FIG.2h). Presumably, antiviral AGOs cannot
resolve such an extensively base-paired RNA, a notion
that is also applicable to the highly-structured 8S RNA
leader of CaMV, whose DCL-mediated processing gen-
erates vsiRNAs that are comparable in abundance to the
entire complement of host-encoded siRNAs and miR-
NAs81. These products target PTGS against the 8S leader
inefficiently, however, and thus probably act mainly as
decoys diverting the silencing machinery from the main
viral transcripts. In a non-mutually exclusive possibility,
some 8S leader-derived vsiRNAs might be hijacked to
target host, as opposed to viral, transcripts82, a notion
that has received further attention notably in the context
of viral symptom development (BOX3).
Emerging themes and considerations in antiviral
silencing. Historically, the work on VSRs arose from
experiments that were designed to understand viral
pathogenicity and virulence55–57. Given that there are
probably as many VSRs as there are plant viruses, the
widespread availability of cloned viral genomes makes
the identification of novel VSRs an almost trivial task
nowadays. This task is greatly facilitated by the existence
of a standardized transient expression assay in tobacco
leaves, in which a reporter transgene, usually GFP, is
both an inducer and a target of PTGS (FIG.3a), such
that co-expression with a VSR dramatically enhances
GFP accumulation83. An alternative, or often comple-
mentary, approach relies on constitutive expression of
isolated VSRs in plants (usually A.thaliana) exhibiting
Figure 2 | Examples of silencing suppression and evasion mechanisms. a | The Turnip
crinkle virus protein P38 binds and inhibits ARGONAUTE 1 (AGO1) function through
Gly-Trp (GW) AGO-hook residues. b | AGO1 ubiquitylation is induced by polerovirus
P0 and P25 from Potato virus X, leading to its degradation through autophagy or the
activity of the 26S proteasome, respectively. c | The Cauliflower mosaic virus-encoded
transactivator protein (TAV) requires an intact nuclear localization signal (NLS) to bind
DOUBLE-STRANDED-RNA-BINDING PROTEIN 4 (DRB4) in the nucleus and thereby
reduce small interfering RNA (siRNA) production. d | V2 of Tomato yellow leaf curl virus
binds the RNA-DEPENDENT RNA POLYMERASE 6 (RDR6) cofactor SUPPRESSOR OF
GENE SILENCING 3 (SGS3) to prevent denovo double-stranded RNA synthesis. e | The
Tobacco mosaic virus (TMV)-encoded 126-kDa replicase subunit (P126) prevents the
methylase activity of HUA ENHANCER1 (HEN1), leading to small RNA uridylation and
degradation. f | RNA endoribonuclease III (RNaseIII) of Sweet potato chlorotic stunt virus
degrades small RNAs into 14-nucleotide (nt) long, inactive species. g | The tombusviral
protein P19 binds and sequesters small RNAs in the 21-nt size range to prevent their
activity in AGO proteins but is selectively unable to associate with miR168, which
represses the AGO1 mRNA, thereby depleting the cellular pool of antiviral AGO1.
h | Potato spindle tuber viroid (PSTVd) gives rise to abundant Dicer-like (DCL)-dependent
siRNAs but is resistant to targeting by AGO owing to the compact structure of its
covalently closed, complementary RNA genome.
REVIEWS
750
|
NOVEMBER 2013
|
VOLUME 11 www.nature.com/reviews/micro
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved
Nature Reviews | Microbiology
N. tabaccum
N. debneyi
GUGAGAGAUGCAGAGCUGAGA
UACUCUUUACGUCUCGACUUU
5′- -3′
3′-
CHLI
mRNA
Y-Sat RNA
CHLI
mRNA
Y-Sat RNA
-5′
GUGAGAGAUGCAGAGCUUAGA
UACUCUUUACGUCUCGACUUU
5′- -3′
3′- -5′
transgene PTGS84–86 (FIG.3b). Although both approaches
are certainly necessary, they are far from being suffi-
cient to reach a conclusion on the VSR status of can-
didate proteins or their modes of action. For instance,
Escherichia coli RNase III efficiently suppresses GFP
silencing in the transient assay, as do other dsRNA-
binding proteins, including some factors isolated from
mammalian viruses87,88, but it would be far-fetched to
conclude that any of these proteins has evolved primarily
to suppress RNA silencing during authentic infection.
Stable VSR expression in silenced transgenic plants can
also be misleading because the output of suppression
will be influenced by the mere nature of the silencing
system used. Hence, VSRs acting specifically at the
level of silencing amplification (for example, RYSV P6
or TYLCV V2) will not be identified in silencing sys-
tems triggered by RNA fold-back transgenes, which are
intrinsically RDR6 and SGS3 independent68,69,89.
These considerations, along with others detailed
below, point to the necessity of studying VSRs in cognate
infection contexts: that is, as part of their viral genomes
of origin. Although complicated by the compact nature
of viral genomes, engineering viral clones with VSR
reporter-fusion or mutant alleles offers the respective
advantages of providing an accurate view of the level,
timing and spatial distribution of VSR action during
bonafide host–pathogen interactions and of allowing
a precise appreciation of the constraints imposed by the
host silencing machinery on distinct phases of virus biol-
ogy (FIG.3c). For instance, only by specifically disabling
expression of the gene encoding P19 (which is nested
within the P22 ORF) was it possible to appreciate that
the main constraint to systemic accumulation of the
resulting P19-deficient tombusvirus was at the level of its
phloem unloading90. Movement of tombusvirus-derived
siRNAs from the phloem to adjacent cells, a process that
is normally precluded by their P19-mediated sequestra-
tion, presumably mediated immunity in recipient cells.
Incidentally, these indirect observations provide some
of the most convincing evidence to date to support a
role for vsiRNA movement in plant antiviral silencing.
The use of viruses carrying inactive VSR alleles is also
instrumental in deciphering the genetics of antiviral
RNA silencing. Hence, wild-type A.thaliana is immune
to infection by P38-deficient TCV, HcPro-deficient
TuMV or 2b-deficient CMV, but this immunity is com-
pletely suppressed by the concurrent loss of DCL2 and
DCL4 function20–22,29,91 (FIG.3c), indicating that func-
tional VSRs normally mask the effects of host silencing
mutations. Therefore, the use of VSR-deficient viruses
emerges as a prerequisite for accurately evaluating the
contributions of distinct silencing components to limit-
ing virus replication, cell-to-cell and systemic spread or
even plant-to-plant transmission. Although currently
restricted to a handful of viruses, this methodology
holds great promise, particularly when combined with
host gene replacement strategies, as exemplified with
ago2 knockout A.thaliana engineered with a catalytic-
null AGO2 allele: the failure of these plants to contain
infection by HcPro-deficient TuMV demonstrates
unambiguously the key contribution of AGO2-mediated
slicing to antiviral silencing31.
The access to stable loss-of-function VSR alleles is
therefore a necessity in contemporary plant antiviral
silencing studies, an endeavour that is potentially facili-
tated by the small size of most VSRs, which makes them
amenable to saturation mutagenesis. This approach was
successful in identifying key amino acids of the PVX-
encoded P25 protein, which, when mutated, uncou-
pled its role in viral movement from its VSR function92.
Comparative genomic analysis of related viral strains
with contrasted virulence properties can also identify
Box 3 | Do plant viruses hijack RNA silencing?
Mammalian viruses can modify host silencing in similar ways to those described in
plants161,162, and some DNA viruses encode their own suite of discrete microRNAs
(miRNAs) that are processed and operated by the vertebrate host cell miRNA
machinery163. These molecules are used both in cis to modulate viral genome
expression and in trans to target host mRNAs involved in immune responses, allowing
viruses to evade such responses and establish long-lasting, latent infections164–167.
Unlike their mammalian counterparts, plant viruses are not known to produce
dedicated miRNAs, but they often modify the miRNA pathway of their hosts owing to
the action of their viral suppressors of RNA silencing (VSRs)168,169. This causes
developmental symptoms, such as the typical leaf serration phenotype triggered by
suppression of miR167-mediated regulation of AUXIN RESPONSE FACTOR 8 (ARF8)170.
Whether the underlying physiological changes benefit viruses remains unclear, but
Turnip mosaic virus replication was not altered in arf8 mutants, despite mimicking
miR167 overexpression.
Given the sheer amount of virus-derived small interfering RNAs (vsiRNAs)
accumulating generally in infected plants and the relative mismatch tolerance
displayed by siRNA–target pairs, the possibility exists that some vsiRNAs might share
sufficient complementarity with host transcripts to promote their silencing. This was
indeed demonstrated with discrete vsiRNAs produced on infection of Arabidopsis
thaliana with Cauliflower mosaic virus (CaMV)82 and a crucifer strain of Tobacco mosaic
virus (TMV-Cg)50, and by two grapevine-infecting viruses38, suggesting that host mRNA
silencing is common to virus infection. However, the significance for pathogenicity
remains unclear, as individual ablation of vsiRNA targets had no significant impact on
CaMV or TMV-Cg infections. The idea that some vsiRNAs might functionally target host
transcripts is also generally at odds with the ability of most phytoviruses to actively
suppress silencing, unless VSR expression and host mRNA targeting are temporally
and/or spatially separated processes.
If host transcript targeting is favourable to
viruses, it might also be expected to arise often
by positive selection of some vsiRNAs. This can
be exemplified by the disease symptoms caused
by the Cucumber mosaic virus (CMV) Y-satellite
RNA (Y-Sat), which relies entirely upon its
helper virus (CMV) for replication and
movement171. Minute nucleotide alterations
to its genome can dramatically alter both
the virulence and host specificity of disease
induction, manifested by typical yellowing
symptoms in Nicotiana tabaccum (tobacco) but
not in the disease-resistant Nicotiana clevelandii
or Nicotiana debneyi (see the figure)137. It was
revealed that a single Y-Sat-derived siRNA
solely accounts for these symptoms by
targeting a partially complementary target site
in a magnesium chelatase subunit mRNA (CHLI),
a key component of the chlorophyll biosynthesis
pathway. Remarkably, the CHLI orthologues
in N.clevelandii and N.debneyi have weaker
pairing sites, preventing the Y-Sat-derived
siRNA effects. Figure modified, with permission,
from REF. 137 © (2011) PLoS.
REVIEWS
NATURE REVIEWS
|
MICROBIOLOGY VOLUME 11
|
NOVEMBER 2013
|
751
FOCUS ON PLANT–MICROBE INTERACTIONS
© 2013 Macmillan Publishers Limited. All rights reserved
Nature Reviews | Microbiology
Systemic
IR transgene P19 transgenic
expression
TCV infection
Inoculated
Unloading
Systemic
Inoculated Primary
lesion
Vasculature
WTdcl2
dcl4
GW
5 dpi 7 dpi
GA
a
c
b
Flower bud
particular residues that are possibly involved in VSR
function, as virulence is often — but not always —
linked to silencing suppression66. Finally, VSR point-
mutant alleles are particularly useful in linking structural
and/or biochemical properties of these proteins to their
silencing-antagonistic functions invivo; siRNA binding,
in particular, is a feature of many VSRs invitro93 that is
rarely validated by studies of siRNA binding-deficient
VSR alleles inplanta.
PTGS and non-viral plant pathogens
Nearly 1,000 miRNA genes (BOX1) are found in A.thali-
ana, and many more might populate larger plant
genomes94,95. As a substantial fraction of miRNAs
responds to biotic and abiotic stress, it comes as no sur-
prise that the induction or repression of specific miRNAs
contributes to the defence of plants against a broad
range of non-viral pathogens96–98. The rapid and selec-
tive downregulation of miR156 on infection of wheat
by the fungal pathogen Erysiphe graminis, for instance,
is accompanied by a reciprocal increase in steady-state
mRNA levels of two validated miR156 targets, including
Ta3711 (REF.99). Ta3711 encodes a putative transcrip-
tion factor of the SQUAMOSA PROMOTER-BINDING
LIKE (SPL) protein family, some members of which
positively regulate plant defence responses (see also a
later section of this Review). The downregulation of 10
out of 11 large miRNA families of loblolly pine (Pinus
taeda) — including seven pine-specific families — was
also observed in response to infection with the rust fun-
gus Cronartium quercuum100. Most validated targets of
the pine-specific families include receptor-like kinases
Figure 3 | Methods to investigate antiviral RNA silencing in plants. a | Classic ‘patch assay’ experiment used to explore
silencing suppression. Co-infiltrating Nicotiana benthamiana leaves with Agrobacterium tumefaciens strains conferring
GFP expression and P38 expression promotes strong green fluorescence in the delivered area if the Argonaute
(AGO)-hook resides (Gly-Trp; GW) are present in P38. This is not observed, however, if the experiment involves a stable,
non-functional P38 allele carrying a mutation in the AGO-hook residues (shown as Gly-Ala; GA) (dashed region). In this
case, RNA silencing rapidly suppresses transient GFP expression. This assay can be useful, but is not sufficient, for
preliminary studies of candidate viral suppressors of RNA silencing (VSRs) and mutant alleles thereof. b | Use of silencing
reporter lines. In the Arabidopsis thaliana SUC–SUL transgenic system, an inverted-repeat (IR) double-stranded RNA
expressed in the vasculature is processed into small interfering RNAs directing non-cell autonomous post-transcriptional
gene silencing of the SULPHUR transcript, in turn causing vein-centred chlorosis (left). SUL silencing is suppressed in
plants co-expressing a second, P19-encoding transgene (middle), which causes typical leaf serrations resulting from
perturbed microRNA action. Turnip crinkle virus (TCV) also suppresses SUL silencing in systemically infected leaves owing
to P38 expression (right). c | An infectious virus in which red fluorescent protein (RFP) replaces the VSR can be used to
investigate the dynamics of antiviral silencing in whole A.thaliana plants. The mutant virus is unable to infect wild-type
(WT) A.thaliana (right) but causes strong symptoms in dicer-like 2 (dcl2) dcl4 double mutant plants (left) because the lack
of antiviral silencing in the host compensates for the lack of VSR in the virus. The use of a fluorescent tag and distinct host
mutant backgrounds allows a precise assessment of the various steps at which silencing affects virus biology, including in
primary lesion development, vascular transport and systemic unloading.
REVIEWS
752
|
NOVEMBER 2013
|
VOLUME 11 www.nature.com/reviews/micro
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved
Nature Reviews | Microbiology
Secretory
vesicle
PR1
AvrPtoB
AvrPto
HopT1-1
miR393 miR393*
AAA
TIR1
mRNA
MIR393
T3SS
Pst FLS2
?
AGO2
AGO2
AGO2
MEMB12
mRNA
PR1 secretory
pathway via
cis
-Golgi
Basal
defence
Auxin
TIR1
MEMB12
AAA
AGO1
N
H
OH
O
hrcC
A mutant of Pseudomonas
syringae pv. tomato that lacks
a functional typeIII secretion
system, thus making it unable
to deliver virulence effectors
into its host.
TypeIII secretion system
(T3SS). A delivery system
encoded by a wide range of
Gram-negative bacterial
pathogens of plants and
animals that is used to inject
effector proteins into host cells.
Avirulent
A pathogen that elicits a
specific effector-triggered
immunity response owing to
recognition of one of its
virulence factors by a cognate
host-encoded resistance gene,
resulting in an incompatible
interaction.
that are possibly involved in PTI and R proteins involved
in ETI (BOX2). Nearly all of these protein targets accu-
mulated to high levels in healthy tissues surrounding
the disease, suggesting that miRNA downregulation
around the infected zone activates defence-related genes
to restrict fungal growth. The downregulation of miR-
NAs controlling R gene transcripts was also observed in
soybean infected with the oomycete Phytophthora sojae,
whereas other miRNA families were induced101. Thus,
a recurrent regulatory pattern emerges, whereby com-
ponents of both ETI and PTI seem to be constitutively
repressed by specific miRNAs in non-infected tissues but
activated on fungal- or oomycete-induced repression of
these miRNAs (see below).
A distinct pattern of miRNA-mediated gene regula-
tion was uncovered in interactions between A.thaliana
and the bacterial pathogen Pseudomonas syringae pv.
tomato (Pst), the best-characterized non-viral disease in
the context of RNA silencing so far. A.thaliana treatment
with the PAMP flg22, a conserved peptide derived from
the bacterial flagellum, enhances MIR393 transcription,
as does infection with the hrcC mutant of Pst, which lacks
a functional typeIII secretion system (T3SS) required
for bacterial virulence96,102 (BOX2). miR393 targets the
TRANSPORT INHIBITOR RESPONSE 1 (TIR1) fam-
ily of auxin receptors, which normally promote cell
growth by degrading repressive transcription factors103.
Interestingly, auxin signalling is also activated by viru-
lent pathogens, including Pst, and is known to promote
disease susceptibility104. Thus, flg22-induced miR393
expression contributes to PTI by repressing a suppres-
sor (auxin) of basal defence (FIG.4). The effect of miR393
alone in this potentiation is significant, because consti-
tutive overexpression of a miR393-resistant TIR1 allele
enhances susceptibility to Pst, whereas bacterial growth
is reduced in miR393-overexpressing lines. Nonetheless,
Pst hrcC infection upregulates additional miRNAs unre-
lated to auxin signalling, suggesting a broader contri-
bution of the miRNA pathway to PTI98. Indeed, the
growth and symptoms from the non-virulent Pst hrcC
were significantly restored in hypomorphic dcl1 mutants
defective for miRNA synthesis but not in siRNA path-
way mutants13. Virulent and avirulent Pst also induce the
accumulation of AGO2, as do virulent viruses26,105 (FIG.4).
Moreover, ago2 mutants are more susceptible to infec-
tion with both virulent and avirulent Pst, suggesting that
AGO2 has important roles in antibacterial, in addition to
antiviral, defence, in a parallel role to the highly divergent
AGO1. Indeed, Pst infection stabilizes the accumulation
and loading of several miRNA* (passenger) strands into
AGO2 (REF.105), whereas these molecules are otherwise
rapidly turned over during loading of miRNA–miRNA*
duplexes into AGO1 in uninfected plants (BOX1; FIG.4).
Among the miRNA* loaded into AGO2 is miR393*,
which targets three transcripts involved in membrane
trafficking, including the SNARE protein MEMBRIN12
(MEMB12), a negative regulator of PATHOGENESIS-
RELATED 1 (PR1) secretion. Accordingly, plants carrying
mutations in MEMB12 or overexpressing miR393* dis-
play enhanced secretion of PR1 and increased resistance
to virulent and avirulent Pst105 (FIG.4). Thus, as in the
miR393–TIR1 interaction, miR393* operates by repress-
ing a repressor of defence (MEMB12). Therefore, two
strands from a single Pst-induced miRNA species are
channelled into two separate RNA silencing pathways
to cooperatively increase A.thaliana resistance to Pst.
This is achieved by simultaneously changing the plant
hormonal balance (miR393–AGO1) and by reorganizing
its secretion system (miR393*–AGO2). Studies of other
miRNA–miRNA* pairs that are similarly induced on
infection by Pst or other pathogens should uncover the
extent and importance of this multichannelling process
in plant resistance.
Plant responses to non-viral pathogens also involve
non-miRNA-based PTGS pathways. For instance,
A.thaliana carrying mutations in DCL4, RDR6, SGS3,
AGO7 and specific TGS components (BOX1) were more
susceptible to infection with the fungus Verticillium but
not Fusarium, Botrytis, Alternaria or Plectophaerella,
suggesting that an amplified endogenous siRNA path-
way specifically restricts Verticillium infection106.
Interestingly, in the same study, ago1, hen1 and hasty
Figure 4 | The miR393–miR393* response as a paradigm for antibacterial
silencing. Arabidopsis thaliana infection with Pseudomonas syringae pv. tomato (Pst)
induces accumulation of miR393, which loads into ARGONAUTE 1 (AGO1) to silence the
auxin receptor TRANSPORT INHIBITOR RESPONSE 1 (TIR1); auxin signalling is a negative
regulator of basal defence. Various Pst effectors secreted into host cells through the
typeIII secretion system (T3SS) dampen this silencing response at different stages. In
parallel, AGO2 is induced and loads miR393*, which silences the MEMBRIN 12 (MEMB12)
mRNA; MEMB12 is a SNARE protein that controls part of the secretory pathway through
the cis Golgi. Silencing MEMB12 mRNA causes a change in the secretion pathway that
favours exocytosis of the defence-related PATHOGENESIS-RELATED 1 (PR1) protein.
REVIEWS
NATURE REVIEWS
|
MICROBIOLOGY VOLUME 11
|
NOVEMBER 2013
|
753
FOCUS ON PLANT–MICROBE INTERACTIONS
© 2013 Macmillan Publishers Limited. All rights reserved
SNARE
Soluble NSF
(N-ethylmaleimide-sensitive
factor) attachment protein
(SNAP) receptor, which
regulates secretory vesicle
fusion.
PATHOGENESIS-RELATED 1
(PR1). A secreted protein with
suspected antimicrobial
functions. It is used as a classic
defence marker owing to its
strong induction during
pathogen challenge.
Hypersensitive response
(HR). A specific form of
programmed cell death, often
induced by effector-triggered
immunity and correlated with
accumulation of antimicrobial
compounds and systemic
acquired resistance.
Bacterial suppressors of
RNA silencing
(BSRs). Effector molecules
used by bacterial pathogens to
alter host post-transcriptional
gene silencing or, possibly,
transcriptional gene silencing.
mutants (BOX1) displayed reduced fungal growth, sug-
gesting that a successful infection relies on the integrity
of the host miRNA pathway, revealing unsuspected
complexity in the A.thaliana–Verticillium interaction.
Other examples of siRNA-mediated defence responses
include the natural antisense (NAT) RNA nat-siR-
NAATGB2, which is induced specifically during ETI
against avirulent Pst. nat-siRNAATGB2 downregulates
a constitutively expressed pentatricopeptide repeat pro-
tein (PPRL) mRNA in cis, in turn facilitating the onset of
the hypersensitive response (HR)107 (BOX2). A second RNA
induced during ETI belongs to a class of siRNAs termed
long siRNAs (lsiRNAs); these siRNAs are atypical in size
(39–41 nt long) and accumulate in biotically stressed
A.thaliana108. The contribution of nat-siRNAATGB2
and lsiRNA-1 to AvrRpt2-specified ETI is suggested
by the fact that mutations in RDR6 and HYPONASTIC
LEAVES 1 (HYL1), genes that are required for their bio-
genesis, compromise RESISTANT TO P. SYRINGAE 2
(RPS2)-mediated resistance. Because more than 30% of
the A.thaliana genome can produce transcripts from
both sense and antisense strands109, it is possible these
examples are merely a glimpse of a more general contri-
bution of nat-siRNA or lsiRNAs to ETI and PTI against
a wide range of pathogens.
PTGS suppression by non-viral plant pathogens.
Differential growth of Pst on A.thaliana silencing
mutants was barely discernible in infections involving
fully virulent bacteria, initially suggesting that PTGS
does not have a significant role in antibacterial immu-
nity13,107. Analogous to the confounding effect of VSRs
during virulent virus infections, the gain in bacterial
growth and symptoms in miRNA-deficient A.thali-
ana only became clearly apparent following infections
with the hrcC T3SS mutant of Pst13 (BOX2), suggesting
that some of the virulence factors injected by Pst into
host cells, also known as ‘effectors’, have evolved to sup-
press PTGS. This notion was effectively validated by
the identification of at least three bacterial suppressors
of RNA silencing (BSRs, by analogy to VSRs) as part of
the Pst ‘effectorome’ (FIG.4). Transgenically expressed
AvrPtoB suppressed the transcriptional activation of
PAMP-responsive A.thaliana miRNA genes, such as
MIR393, whereas AvrPto inhibited miRNA accumu-
lation post-transcriptionally, possibly by impairing
miRNA processing by DCL1 in a manner that required
AvrPto plasma membrane localization. A third bacte-
rial effector, HopT1-1, inhibited cleavage and/or trans-
lational repression of several endogenous miRNA targets
in transgenic A.thaliana, mimicking the effects of ago1
mutants13 (FIG.4).
Recently, the first silencing suppressors from a eukary-
otic plant pathogen were isolated from the soybean-
infecting oomycete P. s o ja e using primarily the tobacco
leaf transient assay evoked in an earlier section of this
Review14. The effects of these Phytophthora-encoded
suppressors of RNA silencing (PSRs) were confirmed on
their transgenic expression in A.thaliana, in which PSR1
and PSR2 affected distinct endogenous PTGS pathways.
Importantly, downregulating PSR2 expression in P. s o j a e
reduced its virulence, whereas infection of tobacco leaves
was enhanced upon transient co-expression of several
VSRs, unravelling a crucial role for silencing suppression
in oomycete infection. Given the broad involvement of
endogenous sRNAs in biotic stress responses, it is antici-
pated that effector proteins secreted into host cells by
fungi, nematodes or insects will be soon additionally
identified as silencing suppressors.
TGS and plant pathogens
Plant DNA viruses, which comprise geminiviruses and
pararetroviruses, replicate as nuclear minichromosomes
or episomes, the expression of which is targeted by both
PTGS and TGS110,111. A role for TGS in defence against
DNA virus infection is exemplified by the hypersuscepti-
bility to geminivirus infection that is displayed by various
A.thaliana RNA-directed DNA methylation (RdDM)
pathway mutants112 (BOX1). Accordingly, geminivirus
genomes accumulate DNA methylation and repressive
histone marks that are typical of RdDM113. The same
DCL3-dependent mechanisms that normally target
RdDM to repeats, transposons and transgenes in plant
genomes (BOX1) are likely to account for the silencing of
minichromosomes and episomes by the 24-nt vsiRNAs
that are formed during DNA virus infections. Indeed,
recombinant gemini- and pararetroviruses can induce
denovo cytosine methylation and TGS of homologous
transgenes stably integrated into the A.thaliana genome,
a dynamic process that is paralleled by a decrease in
disease severity114,115.
Plant DNA viruses have evolved ways to counteract
both TGS and PTGS pathways, and several examples
of DNA virus-encoded PTGS suppressors were pro-
vided above. TGS suppression has been mostly docu-
mented with geminivirus-encoded VSRs, the effects of
which converge on modulating the general availability
or transferability of methyl groups by targeting key
enzymes in the S-adenosyl methionine (SAM) path-
way116–119. More recently, a parallel counter-defensive
strategy was discovered, whereby the replication-asso-
ciated protein (Rep) of several geminiviruses reduces
the expression of METHYLTRANSFERASE 1 (MET1)
and CHROMOMETHYLASE 3 (CMT3), which are
involved in DNA replication-coupled methylation,
at symmetrical CG and CHG sites, respectively120.
Accordingly, Rep expression in A.thaliana was suf-
ficient to suppress methylation of a reporter transgene
undergoing MET1-dependent TGS. It is striking
that geminiviruses suppress TGS by targeting general
enzymatic reactions (the SAM pathway) or DNA repli-
cation-coupled steps, which are respectively peripheral
or dispensable to RdDM. This strategy may have been
driven by the fact that direct suppression of RdDM
pathway components seems to promote plant defences,
as discussedbelow.
RdDM and plant defence. In addition to PTGS-based
regulation, a role for RdDM and TGS in plant disease
has recently emerged and is rationalized by the now
well-established fitness cost incurred by constitutive or
hyperactive defence gene expression in plants121 (BOX2).
REVIEWS
754
|
NOVEMBER 2013
|
VOLUME 11 www.nature.com/reviews/micro
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved
Nature Reviews | Microbiology
WRKY22
RdDM
TE repeat
WRKY22
RdDM
ROS1
TF
SA
O OH
RMG1
RdDM
TE repeat
a b
RMG1
OH
Pst
Me Me Me Me
Me Me
Me Me
Me Me Me
Me Me Me Me
Me Me Me
FLS2 FLS2or
Chromocentres
Densely packed
heterochromatic (silenced)
DNA including pericentromeric
regions, transposons and
repetitive elements.
Salicylic acid
A central plant hormone signal
that induces local and systemic
defence responses, including
systemic acquired resistance.
Biotrophic
Organisms that complete their
life cycle on living plant tissue;
such organisms actively
prevent host cell death.
Necrotrophic
Organisms that kill host cells
before invasion and gain
nutrients from the dead plant
tissue.
Jasmonic acid
A phytohormone that regulates
defence against necrotrophic
pathogens and herbivores.
Dampening defence gene expression through active
RdDM could provide an effective mode of regulation
because RdDM can be rapidly reversed by biotic and abi-
otic stress. For instance, Pst infection is known to induce
rapid DNA hypomethylation at pericentromeric repeats
and decondensation of chromocentres in A.thaliana122.
The rapid activation of plant defences would require
the presence of RdDM-prone genomic segments in the
vicinity of defence-related genes (for example, transpo-
sons and repeats) (BOX1) and the involvement of active
demethylation pathways to ensure optimal and rapid
defence gene induction on pathogen attack. A third
anticipated attribute would be the capacity of the sys-
tem to resume RdDM and thereby suppress defence
gene activation after the stress has elapsed. Profiling
the entire A.thaliana DNA methylome indeed revealed
that many genomic regions enriched in transposon
sequences become differentially methylated on infec-
tion by virulent and avirulent Pst or on salicylic acid
treatment. Moreover, many of these epigenetic altera-
tions affect expression of neighbouring protein-coding
genes, including defence-related genes123.
A separate study revealed that A.thaliana treatment
with flg22 causes a rapid and transient downregula-
tion of key RdDM pathway components, including
AGO4 and the RNA polymerase IV (Pol IV) subunit
NUCLEAR RNA POLYMERASE D 1A4 (BOX1). This
downregulation occurred as early as 3hours post-treat-
ment and was sufficient to reactivate a transposon-based
reporter transgene undergoing TGS. Flg22 also triggered
the rapid DNA demethylation and transcriptional reacti-
vation of several well-characterized endogenous RdDM
target loci in a manner facilitated by the DNA glycosylase
REPRESSOR OF SILENCING 1 (ROS1), which is the
main denovo demethylase in vegetative tissues. These
effects were significant in mediating defence against Pst
because bacterial growth was impaired by mutations
in domains rearranged methyltransferase 1 (drm1) and
drm2, which encode the two A.thaliana denovo meth-
yltransferases recruited by AGO4 and 24-nt rasiRNAs
during RdDM (BOX1; FIG.5); conversely, bacterial growth
was enhanced in ros1 mutant A.thaliana. Wild-type
plants treated with flg22 resembled non-treated RdDM
mutants, in that they exhibited enhanced PR1 expression
and a spontaneous HR4. A separate study showed that
mutating the RNA Pol V complex (BOX1) also increases
plant resistance to biotrophic pathogens such as Pst by
enhancing salicylicacid signalling, a common character-
istic of RdDM mutants124; pol V mutants were, however,
less resistant to necrotrophic fungal pathogens, which are
normally restricted by jasmonic acid, a compound whose
effects are antagonized by salicylic acid.
The enhanced PR1 expression, spontaneous HR and
increased salicylic acid signalling observed in RdDM
mutants are all indicative of the constitutive activation of
R genes, suggesting that RdDM might negatively regulate
the expression of at least some R genes. RESISTANCE
METHYLATED GENE 1 (RMG1) was identified as one
of these genes in A.thaliana, and its promoter was found
to contain two helitron-derived DNA repeats that are nor-
mally subjected to extensive RdDM in wild-type plants4.
Induction of RMG1 was compromised in ros1 mutant
plants, suggesting that active DNA demethylation of the
RMG1 promoter is part of an induced immune response
(FIG.5), although a role for RMG1 in conferring resistance
to Pst or other pathogens has yet to be established. A
second type of RdDM-dependent regulation was identi-
fied for WRKY22, which also contains a transposon rem-
nant in its promoter and encodes a transcription factor
involved in defence (FIG.5). WRKY22 was significantly
overexpressed in RdDM mutants compared with wild-
type plants, but only in response to flg22. Thus, unlike
for RMG1, which was constitutively activated in naive
RdDM mutants, DNA demethylation merely primed the
transcriptional response of WRKY22 to flg22-responsive
transactivators4 (FIG.5). In effect, the RdDM-mediated
control of WRKY22 expression would not have been
appreciated without the prior activation of PTI by flg22,
which might explain why significant gene expression
changes are seldom observed in RdDM mutant plants
under ideal, stable laboratory conditions125. A key aspect
of the above processes is that the dampening of RdDM
and the resulting defence gene activation occurs only
transiently. For instance, DNA methylation and TGS
were already restored to previous levels 9hours after
flg22 treatment of wild-type A.thaliana. This shows
how, under specific selection pressure, transposon rem-
nants might be co-opted by plant gene promoters to
Figure 5 | Dampening RNA-directed DNA methylation enhances Arabidopsis
thaliana resistance against Pseudomonas syringae. Both WRKY22 and RESISTANCE
METHYLATED GENE 1 (RMG1) have transposon element (TE) remnants in the vicinity of
their promoters that attract the RNA-directed DNA methylation (RdDM) machinery. a | In
the absence of infection, DNA methylation precludes (WRKY22) or dampens (RMG1)
transcription. b | Pathogen-associated molecular pattern-triggered immunity (PTI)
activation by Pseudomonas syringae pv. tomato (Pst) or treatment with salicylic acid (SA)
downregulates the RdDM pathway through repression of RNA polymerase IV (Pol IV)
and ARGONAUTE 4 (AGO4) and induction of REPRESSOR OF SILENCING 1 (ROS1)
demethylase activity, resulting in transient differentially methylated regions. This
promotes direct transcriptional activation of RMG1 and priming of WRKY22, which
requires one or several PTI-induced transcription factors (TFs) for its transcriptional
activity. In additional to other loci, RMG1 activation probably underpins the constitutive
defence activation seen in some RdDM mutants. FLS2, FLAGELLIN-SENSITIVE 2.
REVIEWS
NATURE REVIEWS
|
MICROBIOLOGY VOLUME 11
|
NOVEMBER 2013
|
755
FOCUS ON PLANT–MICROBE INTERACTIONS
© 2013 Macmillan Publishers Limited. All rights reserved
Helitron
A DNA transposon that is
commonly subjected to
repressive DNA methylation
and that transposes through
rolling circle DNA replication.
Extreme resistance
A plant defence response
against viruses that is
elicited by effector-triggered
immunity. It differs from the
hypersensitive response in that
it restricts viral accumulation
without cell death.
confer dynamic and stress-responsive expression in ways
that prevent their prolonged induction. This feature is
foreseeably advantageous in the case of defence-related
genes whose continuous expression reduces plant fitness.
Because transposons and their remnants display largely
stochastic insertional patterns, extensive natural varia-
tion might exist among different A.thaliana ecotypes
to facilitate their defence against various pathogens.
Transposons may also have similar and perhaps more
extensive roles in larger, transposon-rich plant genomes,
including those of crop species.
Host counter-counter defence
PTGS control of R gene arrays. The widespread occur-
rence of RNA silencing suppression prompts the ques-
tion of how host plants might adequately respond to the
alteration of their silencing pathways by pathogens. These
counter-counter defence responses are not only antici-
pated to occur on the host side to neutralize the effects of
VSRs, BSRs and PSRs, but also to maintain, on the patho-
gen side, a dynamic selection pressure underpinning the
continuous emergence of new suppressor alleles in micro-
bial populations. An apparently widespread mechanism
was recently uncovered, whereby arrays of related R
genes undergo synchronous post-transcriptional control
by endogenous DCL4-dependent siRNAs10–12,126 (FIG.6).
In addition to the TGS regulation of some R genes (for
example, RMG1), this system not only contributes to the
prevention of the fitness costs of over-active resistance but
also promotes the upregul ation of R gene transcripts when
PTGS is compromised by pathogens, effectively exploit-
ing R gene-derived siRNAs as molecular sensors of infec-
tion. The mechanism may provide an optimal temporal
window for recognition of specific pathogen effectors by
dedicated R genes, the expression of which would thereby
coincide precisely with the virulent phase of infection
(BOX2; FIG.6). Secondly, given that R gene overexpression
causes constitutive defence activation in the absence of
pathogens121,127,128, the simultaneous, bulk expression of
many R genes as a result of PTGS suppression would be
likely to increase host resistance even without specific rec-
ognition of pathogen-derived avirulence factors. PTGS-
mediated control of R genes was first demonstrated for
the A. thaliana RECOGNITION OF PERONOSPORA
PARASITICA 5 (RPP5) locus, which contains the R genes
RPP4 and SUPPRESSOR OF NPR1-1, CONSTITUTIVE 1
(SNC1)129. Members of this cluster are repressed by 21-nt
siRNAs generated by antisense transcription from SNC1.
SNC1 transcripts are upregulated in plants compromised
in siRNA-mediated PTGS, notably dcl4 and ago1, two
mutations phenocopied by VSR expression. Thus, viruses
might functionally upregulate RPP5-resident gene expres-
sion, as might some bacteria and oomycetes, which also
suppress PTGS by effector delivery. Accordingly, most
RPP5 genes are paralogous duplications encoding defence
proteins against these two classes of microorganisms.
The RPP5 example demonstrates the PTGS-mediated
control of genetically clustered R gene variants in cis, but
more recent findings show how trans-acting sRNA pop-
ulations coincidently regulate many genetically unlinked
R genes. First discovered in the legume Medicago
truncatula, these ‘anti-R gene’ siRNAs are synthesized
by DCL4 from dsRNA amplified by RDR6 following the
cleavage of some R gene transcripts by 22-nt miRNAs, a
scheme similar to that of tasiRNA production10 (BOX1;
FIG.6). The process is also widespread in tomato, potato,
grape, poplar, cotton and gymnosperms but not promi-
nent in A.thaliana, suggesting that its selection may
have been strongest in long-lived perennials or plants
with clonal reproduction with less opportunity to evolve
new R genes through sexual recombination126. Hence, in
potato, validated targets of the 22-nt miR482 include the
Rb, R2 and R3a genes, which mediate resistance against
Phytophthora infestans11. If like P.sojae, P.infestans also
secretes PSRs, their delivery into host cells could result
in quasi-avirulence through the enhanced expression of
an entire R protein suite. In tomato, miR482 recognizes
a conserved sequence encoding the p-loop/walkerA
motif from coiled coil nucleotide-binding and Leu-rich
repeat (CC-NB-LRR) proteins, leading to cleavage of
R gene mRNA and production of secondary siRNAs12.
Importantly, the expression of several of these R gene
transcripts increases on infection with virulent Pst and
various viruses, potentially implicating BSRs and VSRs
in their stimulation. In tobacco, the R gene N, which
confers resistance against TMV, also undergoes regula-
tion bymiR482 (REFS11,130), and N signalling requires
nuclear translocation following activation, which pro-
motes transcriptional reprogramming131. Interestingly,
the SPL6 transcription factor was recently found to con-
tribute to this reprogramming132 and is itself regulated by
miR156. This example shows how R gene responses may
be repressed by sRNAs at multiple points and how, cor-
relatively, silencing suppression by pathogens might pro-
mote upregulation of resistance through multiple targets.
R proteins might monitor VSR, BSR and PSR action.
Given the widespread regulation of PTI and ETI by
sRNAs and the ability of pathogens to suppress silenc-
ing, an ultimate host response to the alteration of
silencing pathways would be to use R proteins to guard
the integrity of key silencing components, a process
whereby VSRs, BSRs and PSRs might effectively trigger
ETI (FIG.6b). Agreeing with this hypothesis, the VSR 2b
from Tomato aspermy virus was found to elicit an HR
when expressed in tobacco by a modified TMV strain133.
Recent work with authentic viral infections revealed
that some Nicotiana species respond to the effects of P19
by activating a highly potent immune response termed
extreme resistance, which protects tissues against Tomato
bushy stunt virus infection in the absence of an HR134.
When delivered non-virally (for example, transgeni-
cally) into these species, P19 promotes an HR as well as
salicylate, PR1 and ethylene production characteristic
of ETI. The binding of P19 to sRNAs, which is manda-
tory for its VSR function, is also necessary to induce
this response and, remarkably, addition of unrelated
VSRs that also bind sRNAs specifically suppresses the
onset of the P19-elicited immunity. These results thus
strongly suggest that silencing suppression by P19 can
be sensed in a plant species-specific manner. Although
the reliance of this sensing mechanism on one or several
REVIEWS
756
|
NOVEMBER 2013
|
VOLUME 11 www.nature.com/reviews/micro
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved
Nature Reviews | Microbiology
MIR482a b
Increased
resistance
threshold
Defence
(PTI)
Counter-counter
defence (ETI)
Counter
defence
Infection
VSRsBSRs
or
R1 R2 R3
R1 mRNA
AAA
21nt siRNA
MIR393 vsiRNA
HR threshold
Resistance
threshold
Low
High
Defence amplitude
R
AAA
Viral mRNA
AGO1
AGO1
AGO2
or
vsiRNA
AAA
Viral mRNA
AGO1
AGO1
AGO2
AGO1
R2 mRNA
AAA
AGO1
R3 mRNA
AAA
AGO1
RDR6
SGS3
DCL4
DRB4
R proteins remains to be established, it is interesting to
note that the TCV-encoded VSR, P38, is an avirulence
factor when recognized by the cognate A.thaliana R
protein HRT in specific A.thaliana ecotypes135. This
response triggers an HR-mediated resistance, although
it remains unknown whether the initial recognition
relies on the VSR function of P38. Perhaps more com-
pellingly, the TMV-encoded HR elicitor of the R pro-
tein N has been mapped to the P50 helicase subunit of
the viral replicase, and the same domain is sufficient to
suppress RNA silencing in Nicotiana benthamiana136.
Moreover, the helicase enzymatic activity of P50 is dis-
pensable for both N-mediated resistance and silencing
suppression, suggesting that the VSR activity of P50
might stimulate ETI through the activation of N. It will
be of great interest to determine whether ETI can also
be specifically induced in response to secreted BSRs
andPSRs.
Conclusions and outlook
In addition to its primordial role in antiviral defence, it
is now evident that RNA silencing constitutes a central
regulatory node in plant immunity against a broad range
of pathogens. This involvement is such that, in many
aspects, the schemes of RNA silencing and its suppres-
sion by microorganisms can now be readily superim-
posed on a classic zig-zag-zig model of plant defence9
in the contexts of both virus-derived sRNAs and host-
encoded miRNAs and siRNAs (FIG.6b). In all cases, the
model ultimately implies that dedicated R proteins have
evolved to sense the action of VSRs, BSRs and PSRs, an
idea that is already transpiring from recent studies of
plant–virus interactions. Whether, among the plethora
of plant R proteins, some specifically guard RNA silenc-
ing components is therefore an important future ques-
tion, as is the identity of such R proteins. More generally,
a better understanding of defensive RNA silencing will
Figure 6 | Defence, counter-defence and counter-counter-defence of RNA silencing. a | Constitutive, basal
expression of MIR482 allows the simultaneous silencing of multiple resistance (R) gene family members through
trans-acting small interfering RNAs (tasiRNAs) produced from double-stranded RNA (dsRNA) derived from a primary
miR482-targeted R gene transcript. Pathogen infection and ensuing silencing suppression interferes with the process at
several possible steps, resulting in enhanced accumulation of multiple R proteins and increased defence. b | The classic
zig-zag-zig scheme illustrating: pathogen-associated molecular pattern (PAMP)-triggered immunity (PTI) induction to
raise defence above the threshold for resistance; suppression by pathogen effector molecules to lower defence; and
subsequent triggering of effector-triggered immunity (ETI) through R protein recognition of the cognate effector (or
effectors), leading to hypersensitive response (HR) and strong defence. MIR393 induction by Pseudomonas syringae pv.
tomato (Pst) and suppression of silencing-based R gene control by Pst-encoded effectors (bacterial suppressors of
RNA silencing, BSRs), and virus-derived small interfering RNA (vsiRNA)-mediated induction of RNA silencing and its
suppression by viral suppressors of RNA silencing (VSRs) are readily accommodated in the scheme. AGO, ARGONAUTE;
DCL, DICER-LIKE; DRB4, DOUBLE-STRANDED-RNA-BINDING PROTEIN 4; RDR6, RNA-DEPENDENT RNA
POLYMERASE6.
REVIEWS
NATURE REVIEWS
|
MICROBIOLOGY VOLUME 11
|
NOVEMBER 2013
|
757
FOCUS ON PLANT–MICROBE INTERACTIONS
© 2013 Macmillan Publishers Limited. All rights reserved
entail the clarification of how silencing pathways are
modified by hosts in response to pathogens and by
which specific mechanisms pathogens avoid or manip-
ulate these pathways to ultimately cause disease. In
this respect, the approaches pioneered with VSR stud-
ies can now be adapted to the newly discovered BSRs
and PSRs, which are bound to illuminate novel facets
of PTI and ETI and also improve our understanding of
the host silencing machinery itself. As for VSRs, the key
issues will be to understand the spatial and temporal
dynamics of the action of these proteins in the course
of authentic infections. Also as with VSRs, the genetic
rescue of silencing suppressor-deficient pathogens in
RNA interference-defective hosts and the availability of
stable loss-of-function alleles of effectors will be key to
the future exploration of the contribution of silencing
suppression to virulence.
Although antiviral silencing studies have been at the
forefront of investigations of defensive RNA silencing,
many questions also remain unanswered in this field.
For instance, the popular assumption that vsiRNAs move
throughout plants to convey antiviral immunity still
awaits direct experimental validation, a reminder that
our general understanding of plant antiviral silencing
remains largely based on inferences or educated guesses
made from studies of endogenous-gene or transgene
silencing. The use of dedicated and unbiased forward
genetics may well uncover unexpected and unique facets
of antiviral RNA silencing, including interconnections
with hormone-based or protein-based immune path-
ways, with general RNA metabolism or with factors
important for subcellular dynamics and organization.
The example of the Y-satellite RNA (Y-Sat)-mediated
silencing of the host gene CHLI (a magnesium chelatase
subunit) (BOX3) also shows how subtle effects that have
evolved through intimate host–pathogen interactions
can control the outcome of infections by modulating dis-
ease symptoms, a poorly understood yet economically
paramount aspect of plant–virus interactions137. As most
of our knowledge on antiviral RNA silencing is based on
infecting A.thaliana with viruses that probably share
only a brief co-evolutionary history with this plant16,
a legitimate question is how many such important
phenomena have been missed during experimentation.
Finally, several recent studies have now demonstrated
that various forms of acquired resistance to pathogens
can be passed on to the offspring through mechanisms
apparently controlled by endogenous siRNAs and
RdDM138–140. Host gene silencing induced by recom-
binant viruses can also be transmitted across genera-
tions under some circumstances and independently of
the inducing virus141. Thus, the fascinating possibility
emerges that defensive RNA silencing, in addition to
its roles in infected plants, also represents a phenotypi-
cally plastic mechanism that might prime resistance in
subsequent generations.
Note added in proof
Recently, Zhu and colleagues demonstrated a require-
ment for DRB4 in stability and function of R proteins,
suggesting another layer connecting the RNA silencing
pathway to ETI (REF. 172).
1. Brodersen,P. & Voinnet,O. The diversity of RNA
silencing pathways in plants. Trends Genet. 22,
268–280 (2006).
2. Sunkar,R., Li,Y.F. & Jagadeeswaran,G. Functions of
microRNAs in plant stress responses. Trends Plant Sci.
17, 196–203 (2012).
3. Carlsbecker,A. etal. Cell signalling by
microRNA165/6 directs gene dose-dependent root
cell fate. Nature 465, 316–321 (2010).
4. Yu,A. etal. Dynamics and biological relevance of DNA
demethylation in Arabidopsis antibacterial defense.
Proc. Natl Acad. Sci. USA 110, 2389–2394 (2013).
5. Voinnet,O. Origin, biogenesis, and activity of plant
microRNAs. Cell 136, 669–687 (2009).
6. Ding,S.W. & Voinnet,O. Antiviral immunity directed
by small RNAs. Cell 130, 413–426 (2007).
7. Zipfel,C. etal. Bacterial disease resistance in
Arabidopsis through flagellin perception. Nature 428,
764–767 (2004).
8. Shan,L. etal. Bacterial effectors target the common
signaling partner BAK1 to disrupt multiple MAMP
receptor-signaling complexes and impede plant
immunity. Cell Host Microbe 4, 17–27 (2008).
9. Jones,J.D. & Dangl,J.L. The plant immune system.
Nature 444, 323–329 (2006).
10. Zhai,J. etal. MicroRNAs as master regulators of the
plant NB‑LRR defense gene family via the production
of phased, trans-acting siRNAs. Genes Dev. 25,
2540–2553 (2011).
11. Li,F. etal. MicroRNA regulation of plant innate
immune receptors. Proc. Natl Acad. Sci. USA 109,
1790–1795 (2012).
12. Shivaprasad,P.V. etal. A microRNA superfamily
regulates nucleotide binding site-leucine-rich repeats
and other mRNAs. Plant Cell 24, 859–874 (2012).
References 10, 11 and 12 uncovered a role for
siRNAs in repressing plant resistance genes
13. Navarro,L., Jay,F., Nomura,K., He,S.Y. & Voinnet,O.
Suppression of the microRNA pathway by bacterial
effector proteins. Science 321, 964–967 (2008).
First report of a silencing suppressor outside of
viruses.
14. Qiao,Y. etal. Oomycete pathogens encode RNA
silencing suppressors. Nature Genet. 45, 330–333
(2013).
First discovery of a silencing suppressor in a
eukaryotic pathogen.
15. Ruiz-Ferrer,V. & Voinnet,O. Roles of plant small RNAs
in biotic stress responses. Annu. Rev. Plant Biol. 60,
485–510 (2009).
16. Pagan,I. etal. Arabidopsis thaliana as a model for the
study of plant-virus co-evolution. Phil. Trans. R.Soc. B
365, 1983–1995 (2010).
17. Burgyan,J. & Havelda,Z. Viral suppressors of RNA
silencing. Trends Plant Sci. 16, 265–272 (2011).
18. Qu,F., Ye,X. & Morris,T.J. Arabidopsis DRB4, AGO1,
AGO7, and RDR6 participate in a DCL4-initiated
antiviral RNA silencing pathway negatively regulated
by DCL1. Proc. Natl Acad. Sci. USA 105,
14732–14737 (2008).
19. Jakubiec,A., Yang,S.W. & Chua,N.H. Arabidopsis
DRB4 protein in antiviral defense against Turnip
yellow mosaic virus infection. Plant J. 69, 14–25
(2012).
20. Deleris,A. etal. Hierarchical action and inhibition of
plant Dicer-like proteins in antiviral defense. Science
313, 68–71 (2006).
21. Garcia-Ruiz,H. etal. Arabidopsis RNA-dependent
RNA polymerases and Dicer-like proteins in antiviral
defense and small interfering RNA biogenesis during
turnip mosaic virus infection. Plant Cell 22, 481–496
(2010).
22. Wang,X.B. etal. The 21-nucleotide, but not
22-nucleotide, viral secondary small interfering RNAs
direct potent antiviral defense by two cooperative
Argonautes in Arabidopsis thaliana. Plant Cell 23,
1625–1638 (2011).
23. Vogler,H. etal. Modification of small RNAs associated
with suppression of RNA silencing by tobamovirus
replicase protein. J.Virol. 81, 10379–10388 (2007).
24. Morel,J.B. etal. Fertile hypomorphic ARGONAUTE
(ago1) mutants impaired in post-transcriptional gene
silencing and virus resistance. Plant Cell 14, 629–639
(2002).
25. Zhang,X., Zhang,X., Singh,J., Li,D. & Qu,F.
Temperature-dependent survival of turnip crinkle
virus-infected Arabidopsis plants relies on an RNA
silencing-based defense that requires DCL2, AGO2,
and HEN1. J.Virol. 86, 6847–6854 (2012).
26. Harvey,J.J. etal. An antiviral defense role of AGO2 in
plants. PLoS ONE 6, e14639 (2011).
27. Jaubert,M.J., Bhattacharjee,S., Mello,A.F.,
Perry,K.L. & Moffett,P. ARGONAUTE2 mediates
RNA silencing antiviral defenses against Potato virus X
in Arabidopsis. Plant Physiol. 156, 1556–1564
(2011).
28. Scholthof,H.B. etal. Identification of an ARGONAUTE
for antiviral RNA silencing in Nicotiana benthamiana.
Plant Physiol. 156, 1548–1555 (2011).
29. Azevedo,J. etal. Argonaute quenching and global
changes in Dicer homeostasis caused by a pathogen-
encoded GW repeat protein. Genes Dev. 24, 904–915
(2010).
References 21, 22 and 29 detail genetic work with
subtle viral mutations that disrupt silencing
suppressors.
30. Akbergenov,R. etal. Molecular characterization of
geminivirus-derived small RNAs in different plant
species. Nucleic Acids Res. 34, 462–471 (2006).
31. Carbonell,A. etal. Functional analysis of three
Arabidopsis ARGONAUTES using slicer-defective
mutants. Plant Cell 24, 3613–3629 (2012).
32. Vaucheret,H. Plant ARGONAUTES. Trends Plant Sci.
13, 350–358 (2008).
33. Laliberte,J.F. & Sanfacon,H. Cellular remodeling
during plant virus infection. Annu. Rev. Phytopathol.
48, 69–91 (2010).
34. Brodersen,P. etal. Isoprenoid biosynthesis is required
for miRNA function and affects membrane association
of ARGONAUTE 1 in Arabidopsis. Proc. Natl Acad.
Sci. USA 109, 1778–1783 (2012).
35. Li,S. etal. MicroRNAs inhibit the translation of target
mRNAs on the endoplasmic reticulum in Arabidopsis.
Cell 153, 562–574 (2013).
References 34 and 35 reveal a functional
association between AGO1 and membranes.
REVIEWS
758
|
NOVEMBER 2013
|
VOLUME 11 www.nature.com/reviews/micro
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved
36. Brodersen,P. etal. Widespread translational
inhibition by plant miRNAs and siRNAs. Science 320,
1185–1190 (2008).
37. Lanet,E. etal. Biochemical evidence for translational
repression by Arabidopsis microRNAs. Plant Cell 21,
1762–1768 (2009).
38. Miozzi,L., Gambino,G., Burgyan,J. & Pantaleo,V.
Genome-wide identification of viral and host
transcripts targeted by viral siRNAs in Vitis vinifera.
Mol. Plant Pathol. 14, 30–43 (2013).
39. Pantaleo,V., Szittya,G. & Burgyan,J. Molecular bases
of viral RNA targeting by viral small interfering RNA-
programmed RISC. J.Virol. 81, 3797–3806 (2007).
40. Hoffer,P. etal. Posttranscriptional gene silencing in
nuclei. Proc. Natl Acad. Sci. USA 108, 409–414
(2011).
41. Kumakura,N. etal. SGS3 and RDR6 interact and
colocalize in cytoplasmic SGS3/RDR6-bodies.
FEBS Lett. 583, 1261–1266 (2009).
42. Mourrain,P. etal. Arabidopsis SGS2 and SGS3 genes
are required for posttranscriptional gene silencing and
natural virus resistance. Cell 101, 533–542 (2000).
43. Schwach,F., Vaistij,F.E., Jones,L. & Baulcombe,D.C.
An RNA-dependent RNA polymerase prevents
meristem invasion by potato virus x and is required for
the activity but not the production of a systemic
silencing signal. Plant Physiol. 138, 1842–1852
(2005).
44. Gy,I. etal. Arabidopsis FIERY1, XRN2, and XRN3 are
endogenous RNA silencing suppressors. Plant Cell 19,
3451–3461 (2007).
45. Gazzani,S., Lawerson,T., Woodward,D., Headon,R. &
Sablowski,R. A link between mRNA turnover and RNA
interference in Arabidopsis. Science 306, 1046–1048
(2004).
46. Moreno,A.B. etal. Cytoplasmic and nuclear quality
control and turnover of single-stranded RNA modulate
post-transcriptional gene silencing in plants.
Nucleic Acids Res. 41, 4699–4708 (2013).
47. Diaz-Pendon,J.A., Li,F., Li,W.X. & Ding,S.W.
Suppression of antiviral silencing by cucumber mosaic
virus 2b protein in Arabidopsis is associated with
drastically reduced accumulation of three classes of
viral small interfering RNAs. Plant Cell 19,
2053–2063 (2007).
48. Donaire,L. etal. Structural and genetic requirements
for the biogenesis of tobacco rattle virus-derived small
interfering RNAs. J.Virol. 82, 5167–5177 (2008).
49. Wang,X.B. etal. RNAi-mediated viral immunity
requires amplification of virus-derived siRNAs in
Arabidopsis thaliana. Proc. Natl Acad. Sci. USA 107,
484–489 (2010).
50. Qi,X., Bao,F.S. & Xie,Z. Small RNA deep sequencing
reveals role for Arabidopsis thaliana RNA-dependent
RNA polymerases in viral siRNA biogenesis. PLoS ONE
4, e4971 (2009).
51. Dunoyer,P. etal. Small RNA duplexes function as
mobile silencing signals between plant cells. Science
328, 912–916 (2010).
52. Molnar,A. etal. Small silencing RNAs in plants are
mobile and direct epigenetic modification in recipient
cells. Science 328, 872–875 (2011).
53. Montgomery,T.A. etal. Specificity of
ARGONAUTE7-miR390 interaction and dual
functionality in TAS3 trans-acting siRNA formation.
Cell 133, 128–141 (2008).
54. Mi,S. etal. Sorting of small RNAs into Arabidopsis
Argonaute complexes is directed by the 5′ terminal
nucleotide. Cell 133, 116–127 (2008).
55. Brigneti,G. etal. Viral pathogenicity determinants are
suppressors of transgene silencing in Nicotiana
benthamiana. EMBO J. 17, 6739–6746 (1998).
56. Anandalakshmi,R. etal. A viral suppressor of gene
silencing in plants. Proc. Natl Acad. Sci. USA 95,
13079–13084 (1998).
57. Kasschau,K.D. & Carrington,J.C. A counterdefensive
strategy of plant viruses: suppression of
posttranscriptional gene silencing. Cell 95, 461–470
(1998).
58. Voinnet,O., Pinto,Y.M. & Baulcombe,D.C.
Suppression of gene silencing: a general strategy used
by diverse DNA and RNA viruses. Proc. Natl Acad. Sci.
USA 96, 14147–14152 (1999).
59. Schott,G. etal. Differential effects of viral silencing
suppressors on siRNA and miRNA loading support the
existence of two distinct cellular pools of
ARGONAUTE1. EMBO J. 31, 2553–2565 (2012).
60. Baumberger,N., Tsai,C.H., Lie,M., Havecker,E. &
Baulcombe,D.C. The Polerovirus silencing suppressor
P0 targets ARGONAUTE proteins for degradation.
Curr. Biol. 17, 1609–1614 (2007).
61. Csorba,T., Lozsa,R., Hutvagner,G. & Burgyan,J.
Polerovirus protein P0 prevents the assembly of small
RNA-containing RISC complexes and leads to
degradation of ARGONAUTE1. Plant J. 62, 463–472
(2010).
62. Derrien,B. etal. Degradation of the antiviral
component ARGONAUTE1 by the autophagy pathway.
Proc. Natl Acad. Sci. USA 109, 15942–15946
(2012).
63. Gibbings,D. etal. Selective autophagy degrades
DICER and AGO2 and regulates miRNA activity.
Nature Cell Biol. 14, 1314–1321 (2012).
64. Zhang,P. & Zhang,H. Autophagy modulates miRNA-
mediated gene silencing and selectively degrades
AIN-1/GW182 in C.elegans. EMBO Rep. 14, 568–576
(2013).
65. Chiu,M.H., Chen,I.H., Baulcombe,D.C. & Tsai,C.H.
The silencing suppressor P25 of Potato virus X
interacts with Argonaute1 and mediates its
degradation through the proteasome pathway. Mol.
Plant Pathol. 11, 641–649 (2010).
66. Duan,C.G. etal. Suppression of Arabidopsis
ARGONAUTE1-mediated slicing, transgene-induced
RNA silencing, and DNA methylation by distinct
domains of the Cucumber mosaic virus 2b protein.
Plant Cell 24, 259–274 (2012).
67. Haas,G. etal. Nuclear import of CaMV P6 is required
for infection and suppression of the RNA silencing
factor DRB4. EMBO J. 27, 2102–2112 (2008).
68. Guo,H. etal. Rice yellow stunt rhabdovirus protein 6
suppresses systemic RNA silencing by blocking
RDR6-mediated secondary siRNA synthesis.
Mol. Plant Microbe Interact. 26, 927–936 (2013).
69. Glick,E. etal. Interaction with host SGS3 is required
for suppression of RNA silencing by tomato yellow leaf
curl virus V2 protein. Proc. Natl Acad. Sci. USA 105,
157–161 (2008).
70. Cuellar,W.J. etal. Elimination of antiviral defense by
viral RNase III. Proc. Natl Acad. Sci. USA 106,
10354–10358 (2009).
71. Ye,K., Malinina,L. & Patel,D.J. Recognition of small
interfering RNA by a viral suppressor of RNA silencing.
Nature 426, 874–878 (2003).
72. Vargason,J.M., Szittya,G., Burgyan,J. & Tanaka
Hall,T.M. Size selective recognition of siRNA by an
RNA silencing suppressor. Cell 115 , 799–811 (2003).
73. Varallyay,E., Valoczi,A., Agyi,A., Burgyan,J. &
Havelda,Z. Plant virus-mediated induction of miR168
is associated with repression of ARGONAUTE1
accumulation. EMBO J. 29, 3507–3519 (2010).
74. Mallory,A.C. & Vaucheret,H. ARGONAUTE 1
homeostasis invokes the coordinate action of the
microRNA and siRNA pathways. EMBO Rep. 10,
521–526 (2009).
75. Mallory,A.C. etal. Redundant and specific roles of
the ARGONAUTE proteins AGO1 and ZLL in
development and small RNA-directed gene silencing.
PLoS Genet. 5, e1000646 (2009).
76. Varallyay,E. & Havelda,Z. Unrelated viral suppressors
of RNA silencing mediate the control of ARGONAUTE1
level. Mol. Plant Pathol. 14, 567–575 (2013).
77. Wu,Q. etal. Homology-independent discovery of
replicating pathogenic circular RNAs by deep
sequencing and a new computational algorithm.
Proc. Natl Acad. Sci. USA 109, 3938–3943 (2012).
78. Flores,R. etal. Viroids: the minimal non-coding RNAs
with autonomous replication. FEBS Lett. 567, 42–48
(2004).
79. Itaya,A. etal. A structured viroid RNA serves as a
substrate for dicer-like cleavage to produce
biologically active small RNAs but is resistant to RNA-
induced silencing complex-mediated degradation.
J.Virol. 81, 2980–2994 (2007).
80. Denti,M.A., Boutla,A., Tsagris,M. & Tabler,M. Short
interfering RNAs specific for potato spindle tuber
viroid are found in the cytoplasm but not in the
nucleus. Plant J. 37, 762–769 (2004).
81. Blevins,T. etal. Massive production of small RNAs
from a non-coding region of Cauliflower mosaic virus in
plant defense and viral counter-defense. Nucleic Acids
Res. 39, 5003–5014 (2011).
82. Moissiard,G. & Voinnet,O. RNA silencing of host
transcripts by cauliflower mosaic virus requires
coordinated action of the four Arabidopsis Dicer-like
proteins. Proc. Natl Acad. Sci. USA 103,
19593–19598 (2006).
83. Voinnet,O., Rivas,S., Mestre,P. & Baulcombe,D.
An enhanced transient expression system in plants
based on suppression of gene silencing by the p19
protein of tomato bushy stunt virus. Plant J. 33,
949–956 (2003).
84. Zhang,X. etal. Cucumber mosaic virus-encoded 2b
suppressor inhibits Arabidopsis Argonaute1 cleavage
activity to counter plant defense. Genes Dev. 20,
3255–3268 (2006).
85. Chapman,E.J., Prokhnevsky,A.I., Gopinath,K.,
Dolja,V.V. & Carrington,J.C. Viral RNA silencing
suppressors inhibit the microRNA pathway at an
intermediate step. Genes Dev. 18, 1179–1186
(2004).
86. Dunoyer,P., Lecellier,C.H., Parizotto,E.A., Himber,C.
& Voinnet,O. Probing the microRNA and small
interfering RNA pathways with virus-encoded
suppressors of RNA silencing. Plant Cell 16,
1235–1250 (2004).
87. Lichner,Z., Silhavy,D. & Burgyan,J. Double-stranded
RNA-binding proteins could suppress RNA
interference-mediated antiviral defences. J.Gen. Virol.
84, 975–980 (2003).
88. Bucher,E., Hemmes,H., de Haan,P., Goldbach,R. &
Prins,M. The influenza A virus NS1 protein binds
small interfering RNAs and suppresses RNA silencing
in plants. J.Gen. Virol. 85, 983–991 (2004).
89. Himber,C., Dunoyer,P., Moissiard,G., Ritzenthaler,C.
& Voinnet,O. Transitivity-dependent and -independent
cell-to-cell movement of RNA silencing. EMBO J. 22,
4523–4533 (2003).
90. Havelda,Z., Hornyik,C., Crescenzi,A. & Burgyan,J.
In situ characterization of Cymbidium ringspot
Tombusvirus infection-induced posttranscriptional
gene silencing in Nicotiana benthamiana. J.Virol. 77,
6082–6086 (2003).
91. Chen,H.Y., Yang,J., Lin,C. & Yuan,Y.A. Structural
basis for RNA-silencing suppression by Tomato
aspermy virus protein 2b. EMBO Rep. 9, 754–760
(2008).
92. Bayne,E.H., Rakitina,D.V., Morozov,S.Y. &
Baulcombe,D.C. Cell-to-cell movement of potato
potexvirus X is dependent on suppression of RNA
silencing. Plant J. 44, 471–482 (2005).
93. Lakatos,L. etal. Small RNA binding is a common
strategy to suppress RNA silencing by several viral
suppressors. EMBO J. 25, 2768–2780 (2006).
94. Fahlgren,N. etal. MicroRNA gene evolution in
Arabidopsis lyrata and Arabidopsis thaliana.
Plant Cell 22, 1074–1089 (2010).
95. Cuperus,J.T., Fahlgren,N. & Carrington,J.C.
Evolution and functional diversification of MIRNA
genes. Plant Cell 23, 431–442 (2011).
96. Navarro,L. etal. A plant miRNA contributes to
antibacterial resistance by repressing auxin signaling.
Science 312, 436–439 (2006).
97. Katiyar-Agarwal,S. & Jin,H. Role of small RNAs in
host-microbe interactions. Annu. Rev. Phytopathol.
48, 225–246 (2010).
98. Fahlgren,N. etal. High-throughput sequencing of
Arabidopsis microRNAs: evidence for frequent birth
and death of MIRNA genes. PLoS ONE 2, e219
(2007).
99. Xin,M. etal. Diverse set of microRNAs are responsive
to powdery mildew infection and heat stress in wheat
(Triticum aestivum L.). BMC Plant Biol. 10, 123 (2010).
100. Lu,S., Sun,Y.H., Amerson,H. & Chiang,V.L.
MicroRNAs in loblolly pine (Pinus taeda L.) and their
association with fusiform rust gall development.
Plant J. 51, 1077–1098 (2007).
101. Guo,N. etal. Microarray profiling reveals microRNAs
involving soybean resistance to Phytophthora sojae.
Genome 54, 954–958 (2011).
102. Li,Y. etal. Identification of microRNAs involved in
pathogen-associated molecular pattern-triggered
plant innate immunity. Plant Physiol. 152,
2222–2231 (2010).
103. Dharmasiri,N., Dharmasiri,S. & Estelle,M. The F-box
protein TIR1 is an auxin receptor. Nature 435,
441–445 (2005).
104. Fu,J. & Wang,S. Insights into auxin signaling in plant-
pathogen interactions. Front. Plant Sci. 2, 74 (2011).
105. Zhang,X. etal. Arabidopsis Argonaute 2 regulates
innate immunity via miRNA393*-mediated silencing
of a Golgi-localized SNARE gene, MEMB12. Mol. Cell
42, 356–366 (2011).
106. Ellendorff,U., Fradin,E.F., de Jonge,R. &
Thomma,B.P. RNA silencing is required for
Arabidopsis defence against Verticillium wilt disease.
J.Exp. Bot. 60, 591–602 (2009).
107. Katiyar-Agarwal,S. etal. A pathogen-inducible
endogenous siRNA in plant immunity. Proc. Natl Acad.
Sci. USA 103, 18002–18007 (2006).
108. Katiyar-Agarwal,S., Gao,S., Vivian-Smith,A. & Jin,H.
A novel class of bacteria-induced small RNAs in
Arabidopsis. Genes Dev. 21, 3123–3134 (2007).
REVIEWS
NATURE REVIEWS
|
MICROBIOLOGY VOLUME 11
|
NOVEMBER 2013
|
759
FOCUS ON PLANT–MICROBE INTERACTIONS
© 2013 Macmillan Publishers Limited. All rights reserved
109. Wang,X.J., Gaasterland,T. & Chua,N.H. Genome-
wide prediction and identification of cis-natural
antisense transcripts in Arabidopsis thaliana.
Genome Biol. 6, R30 (2005).
110 . Raja,P., Wolf,J.N. & Bisaro,D.M. RNA silencing
directed against geminiviruses: post-transcriptional
and epigenetic components. Biochim. Biophys. Acta
1799, 337–351 (2010).
111. Chellappan,P., Vanitharani,R. & Fauquet,C.M. Short
interfering RNA accumulation correlates with host
recovery in DNA virus-infected hosts, and gene
silencing targets specific viral sequences. J.Virol. 78,
7465–7477 (2004).
112 . Raja,P., Sanville,B.C., Buchmann,R.C. &
Bisaro,D.M. Viral genome methylation as an
epigenetic defense against geminiviruses. J.Virol. 82,
8997–9007 (2008).
113 . Dogar,A.M. RNAi dependent epigenetic marks on a
geminivirus promoter. Virol. J. 3, 5 (2006).
114 . Al-Kaff,N.S. etal. Transcriptional and
posttranscriptional plant gene silencing in response to
a pathogen. Science 279, 2113–2115 (1998).
115 . Seemanpillai,M., Dry,I., Randles,J. & Rezaian,A.
Transcriptional silencing of geminiviral promoter-
driven transgenes following homologous virus
infection. Mol. Plant Microbe Interact. 16, 429–438
(2003).
116 . Yang,X. etal. Suppression of methylation-mediated
transcriptional gene silencing by βC1-SAHH protein
interaction during geminivirus-betasatellite infection.
PLoS Pathog. 7, e1002329 (2011).
117 . Buchmann,R.C., Asad,S., Wolf,J.N., Mohannath,G.
& Bisaro,D.M. Geminivirus AL2 and L2 proteins
suppress transcriptional gene silencing and cause
genome-wide reductions in cytosine methylation.
J.Virol. 83, 5005–5013 (2009).
118 . Wang,H., Buckley,K.J., Yang,X., Buchmann,R.C. &
Bisaro,D.M. Adenosine kinase inhibition and
suppression of RNA silencing by geminivirus AL2 and
L2 proteins. J.Virol. 79, 7410–7418 (2005).
119 . Zhang,Z. etal. BSCTV C2 attenuates the degradation
of SAMDC1 to suppress DNA methylation-mediated
gene silencing in Arabidopsis. Plant Cell 23, 273–288
(2011).
120. Rodriguez-Negrete,E. etal. Geminivirus Rep protein
interferes with the plant DNA methylation machinery
and suppresses transcriptional gene silencing.
New Phytol. 199, 464–475 (2013).
121. Tian,D., Traw,M.B., Chen,J.Q., Kreitman,M. &
Bergelson,J. Fitness costs of R-gene-mediated
resistance in Arabidopsis thaliana. Nature 423,
74–77 (2003).
122. Pavet,V., Quintero,C., Cecchini,N.M., Rosa,A.L. &
Alvarez,M.E. Arabidopsis displays centromeric DNA
hypomethylation and cytological alterations of
heterochromatin upon attack by pseudomonas
syringae. Mol. Plant Microbe Interact. 19, 577–587
(2006).
123. Dowen,R.H. etal. Widespread dynamic DNA
methylation in response to biotic stress. Proc. Natl
Acad. Sci. USA 109, E2183–E2191 (2012).
With Reference 4, first demonstration that RdDM
represses plant defence.
124. Lopez,A., Ramirez,V., Garcia-Andrade,J., Flors,V. &
Vera,P. The RNA silencing enzyme RNA polymerase v
is required for plant immunity. PLoS Genet. 7,
e1002434 (2011).
125. Henderson,I.R. etal. Dissecting Arabidopsis thaliana
DICER function in small RNA processing, gene
silencing and DNA methylation patterning.
Nature Genet. 38, 721–725 (2006).
126. Kallman,T., Chen,J., Gyllenstrand,N. &
Lagercrantz,U. A significant fraction of 21-nucleotide
small RNA originates from phased degradation of
resistance genes in several perennial species.
Plant Physiol. 162, 741–754 (2013).
127. Palma,K. etal. Autoimmunity in Arabidopsis acd11 is
mediated by epigenetic regulation of an immune
receptor. PLoS Pathog. 6, e1001137 (2010).
128. Bendahmane,A., Kanyuka,K. & Baulcombe,D.C.
The Rx gene from potato controls separate virus
resistance and cell death responses. Plant Cell 11,
781–792 (1999).
129. Yi,H. & Richards,E.J. A cluster of disease resistance
genes in Arabidopsis is coordinately regulated by
transcriptional activation and RNA silencing. Plant Cell
19, 2929–2939 (2007).
130. Whitham,S. etal. The product of the tobacco mosaic
virus resistance gene N: similarity to toll and the
interleukin-1 receptor. Cell 78, 1101–1115 (1994).
131. Burch-Smith,T.M. etal. A novel role for the TIR
domain in association with pathogen-derived elicitors.
PLoS Biol. 5, e68 (2007).
132. Padmanabhan,M.S. etal. Novel positive regulatory
role for the SPL6 transcription factor in the N
TIR-NB-LRR receptor-mediated plant innate immunity.
PLoS Pathog. 9, e1003235 (2013).
133. Li,H.W. etal. Strong host resistance targeted against
a viral suppressor of the plant gene silencing defence
mechanism. EMBO J. 18, 2683–2691 (1999).
134. Sansregret,R. etal. Extreme resistance as a host
counter-counter defense against viral suppression of
RNA silencing. PLoS Pathog. 9, e1003435 (2013).
Demonstrates a strong defence response triggered
specifically by the activity of a VSR.
135. Cooley,M.B., Pathirana,S., Wu,H.J., Kachroo,P. &
Klessig,D.F. Members of the Arabidopsis HRT/RPP8
family of resistance genes confer resistance to both
viral and oomycete pathogens. Plant Cell 12, 663–676
(2000).
136. Wang,L.Y. etal. Multiple domains of the tobacco
mosaic virus p126 protein can independently
suppress local and systemic RNA silencing. Mol. Plant
Microbe Interact. 25, 648–657 (2012).
137. Smith,N.A., Eamens,A.L. & Wang,M.B. Viral small
interfering RNAs target host genes to mediate disease
symptoms in plants. PLoS Pathog. 7, e1002022
(2011).
Clear demonstration of disease symptoms caused
by a vsiRNA targeting a host gene.
138. Rasmann,S. etal. Herbivory in the previous
generation primes plants for enhanced insect
resistance. Plant Physiol. 158, 854–863 (2012).
139. Slaughter,A. etal. Descendants of primed
Arabidopsis plants exhibit resistance to biotic stress.
Plant Physiol. 158, 835–843 (2012).
140. Luna,E., Bruce,T.J., Roberts,M.R., Flors,V. & Ton,J.
Next-generation systemic acquired resistance.
Plant Physiol. 158, 844–853 (2012).
141. Kanazawa,A. etal. Virus-mediated efficient induction
of epigenetic modifications of endogenous genes with
phenotypic changes in plants. Plant J. 65, 156–168
(2011).
142. Voinnet,O. Use, tolerance and avoidance of amplified
RNA silencing by plants. Trends Plant Sci. 13,
317–328 (2008).
143. Liu,Q., Feng,Y. & Zhu,Z. Dicer-like (DCL) proteins in
plants. Funct. Integr. Genom. 9, 277–286 (2009).
144. Boutet,S. etal. Arabidopsis HEN1: a genetic link
between endogenous miRNA controlling development
and siRNA controlling transgene silencing and virus
resistance. Curr. Biol. 13, 843–848 (2003).
145. Yu,B. etal. Methylation as a crucial step in plant
microRNA biogenesis. Science 307, 932–935 (2005).
146. Law,J.A. & Jacobsen,S.E. Establishing, maintaining
and modifying DNA methylation patterns in plants
and animals. Nature Rev. Genet. 11, 204–220 (2010).
147. Han,M.H., Goud,S., Song,L. & Fedoroff,N. The
Arabidopsis double-stranded RNA-binding protein
HYL1 plays a role in microRNA-mediated gene
regulation. Proc. Natl Acad. Sci. USA 101, 1093–1098
(2004).
148. Zilberman,D., Cao,X. & Jacobsen,S.E.
ARGONAUTE4 control of locus-specific siRNA
accumulation and DNA and histone methylation.
Science 299, 716–719 (2003).
149. Kanno,T. etal. Involvement of putative SNF2
chromatin remodeling protein DRD1 in RNA-directed
DNA methylation. Curr. Biol. 14, 801–805 (2004).
150. Zheng,X., Zhu,J., Kapoor,A. & Zhu,J.K. Role of
Arabidopsis AGO6 in siRNA accumulation, DNA
methylation and transcriptional gene silencing.
EMBO J. 26, 1691–1701 (2007).
151. Matzke,M., Kanno,T., Daxinger,L., Huettel,B. &
Matzke,A.J. RNA-mediated chromatin-based
silencing in plants. Curr. Opin. Cell Biol. 21, 367–376
(2009).
152. Chen,H.M. etal. 22-nucleotide RNAs trigger
secondary siRNA biogenesis in plants. Proc. Natl
Acad. Sci. USA 107, 15269–15274 (2010).
153. Cuperus,J.T. etal. Unique functionality of 22-nt
miRNAs in triggering RDR6-dependent siRNA
biogenesis from target transcripts in Arabidopsis.
Nature Struct. Mol. Biol. 17, 997–1003 (2010).
154. Manavella,P.A., Koenig,D. & Weigel,D. Plant
secondary siRNA production determined by
microRNA-duplex structure. Proc. Natl Acad. Sci. USA
109, 2461–2466 (2012).
155. Gomez-Gomez,L. & Boller,T. FLS2: an LRR receptor-
like kinase involved in the perception of the bacterial
elicitor flagellin in Arabidopsis. Mol. Cell 5,
1003–1011 (2000).
156. Chinchilla,D. etal. A flagellin-induced complex of the
receptor FLS2 and BAK1 initiates plant defence.
Nature 448, 497–500 (2007).
157. van der Hoorn,R.A. & Kamoun,S. From guard to
decoy: a new model for perception of plant pathogen
effectors. Plant Cell 20, 2009–2017 (2008).
158. Friedman,A.R. & Baker,B.J. The evolution of
resistance genes in multi-protein plant resistance
systems. Curr. Opin. Genet. Dev. 17, 493–499
(2007).
159. Belkhadir,Y., Nimchuk,Z., Hubert,D.A., Mackey,D.
& Dangl,J.L. Arabidopsis RIN4 negatively regulates
disease resistance mediated by RPS2 and RPM1
downstream or independent of the NDR1 signal
modulator and is not required for the virulence
functions of bacterial typeIII effectors AvrRpt2 or
AvrRpm1. Plant Cell 16, 2822–2835 (2004).
160. Lorrain,S., Vailleau,F., Balague,C. & Roby,D. Lesion
mimic mutants: keys for deciphering cell death and
defense pathways in plants? Trends Plant Sci. 8,
263–271 (2003).
161. Lecellier,C.H. etal. A cellular microRNA mediates
antiviral defense in human cells. Science 308,
557–560 (2005).
162. Lu,S. & Cullen,B.R. Adenovirus VA1 noncoding RNA
can inhibit small interfering RNA and microRNA
biogenesis. J.Virol. 78, 12868–12876 (2004).
163. Cullen,B.R. MicroRNAs as mediators of viral evasion
of the immune system. Nature Immunol. 14, 205–210
(2013).
164. Umbach,J.L. etal. MicroRNAs expressed by herpes
simplex virus 1 during latent infection regulate viral
mRNAs. Nature 454, 780–783 (2008).
165. Sullivan,C.S., Grundhoff,A.T., Tevethia,S.,
Pipas,J.M. & Ganem,D. SV40-encoded microRNAs
regulate viral gene expression and reduce
susceptibility to cytotoxic Tcells. Nature 435,
682–686 (2005).
166. Choy,E.Y. etal. An Epstein–Barr virus-encoded
microRNA targets PUMA to promote host cell survival.
J.Exp. Med. 205, 2551–2560 (2008).
167. Abend,J.R., Uldrick,T. & Ziegelbauer,J.M.
Regulation of tumor necrosis factor-like weak inducer
of apoptosis receptor protein (TWEAKR) expression by
Kaposi’s sarcoma-associated herpesvirus microRNA
prevents TWEAK-induced apoptosis and inflammatory
cytokine expression. J.Virol. 84, 12139–12151
(2010).
168. Lewsey,M., Robertson,F.C., Canto,T., Palukaitis,P. &
Carr,J.P. Selective targeting of miRNA-regulated
plant development by a viral counter-silencing protein.
Plant J. 50, 240–252 (2007).
169. Chen,J., Li,W.X., Xie,D., Peng,J.R. & Ding,S.W.
Viral virulence protein suppresses RNA silencing-
mediated defense but upregulates the role of microrna
in host gene expression. Plant Cell 16, 1302–1313
(2004).
170. Jay,F. etal. Misregulation of AUXIN RESPONSE
FACTOR 8 underlies the developmental abnormalities
caused by three distinct viral silencing suppressors in
Arabidopsis. PLoS Pathog. 7, e1002035 (2011).
171. Wang,M.B. etal. On the role of RNA silencing in the
pathogenicity and evolution of viroids and viral
satellites. Proc. Natl Acad. Sci. USA 101, 3275–3280
(2004).
172. Zhu, S. et al. Double-stranded RNA-binding protein 4
is required for resistance signaling against viral
and bacterial pathogens. Cell Rep. 4, 1168–1184
(2013)
Acknowledgements
Work in the O.V. group is supported by a core grant from
ETH-Z, Advanced Researcher grant ‘Frontiers of RNAi-II’ (No.
323071), from the European Research Council, and a grant
from the Swiss National Foundation (No. 31003A_132907).
N.P. received support from an EMBO Long-Term Fellowship
and a Marie-Curie Actions Incoming International Fellowship.
Competing interests statement
The authors declare no competing financial interests.
FURTHER INFORMATION
Nature Reviews Microbiology Focus on Plant–Microbe
Interactions: http://www.nature.com/nrmicro/focus/plant-
microbe-interactions
ALL LINKS ARE ACTIVE IN THE ONLINE PDF
REVIEWS
760
|
NOVEMBER 2013
|
VOLUME 11 www.nature.com/reviews/micro
REVIEWS
© 2013 Macmillan Publishers Limited. All rights reserved
