Small RNAs containing the pathogenic determinant of a chloroplast-replicating viroid guide the degradation of a host mRNA as predicted by RNA silencing

Small RNAs containing the pathogenic determinant of achloroplast-replicating viroid guide the degradation of a hostmRNA as predicted by RNA silencing

Beatriz Navarro1, Andreas Gisel2, Maria Elena Rodio1, Sonia Delgado3, Ricardo Flores3,* and Francesco Di Serio1,*

1Istituto di Virologia Vegetale (CNR), Unita Organizzativa di Bari, Via Amendola 165/A, 70126 Bari, Italy,2Istituto di Tecnologie Biomediche (CNR), Unita Organizzativa di Bari, Via Amendola 122/D, 70126 Bari, Italy, and3Instituto de Biologıa Molecular y Celular de Plantas (UPV-CSIC), Campus Universidad Politecnica, Avenida de los Naranjos,

46022 Valencia, Spain

Received 27 January 2012; accepted 9 February 2012; published online 2 April 2012.

*For correspondence (e-mail [email protected]; [email protected]).

GenBank accession numbers for the nucleotide sequences reported: JN377825–JN377892.

SUMMARY

How viroids, tiny non-protein-coding RNAs (�250–400 nt), incite disease is unclear. One hypothesis is that

viroid-derived small RNAs (vd-sRNAs; 21–24 nt) resulting from the host defensive response, via RNA silencing,

may target for cleavage cell mRNAs and trigger a signal cascade, eventually leading to symptoms. Peach latent

mosaic viroid (PLMVd), a chloroplast-replicating viroid, is particularly appropriate to tackle this question

because it induces an albinism (peach calico, PC) strictly associated with variants containing a specific 12–14-nt

hairpin insertion. By dissecting albino and green leaf sectors of Prunus persica (peach) seedlings inoculated

with PLMVd natural and artificial variants, and cloning their progeny, we have established that the hairpin

insertion sequence is involved in PC. Furthermore, using deep sequencing, semi-quantitative RT-PCR and RNA

ligase-mediated rapid amplification of cDNA ends (RACE), we have determined that two PLMVd-sRNAs

containing the PC-associated insertion (PC-sRNA8a and PC-sRNA8b) target for cleavage the mRNA encoding

the chloroplastic heat-shock protein 90 (cHSP90), thus implicating RNA silencing in the modulation of host

gene expression by a viroid. Chloroplast malformations previously reported in PC-expressing tissues are

consistent with the downregulation of cHSP90, which participates in chloroplast biogenesis and plastid-to-

nucleus signal transduction in Arabidopsis. Besides PC-sRNA8a and PC-sRNA8b, both deriving from the less-

abundant PLMVd ()) strand, we have identified other PLMVd-sRNAs potentially targeting peach mRNAs.

These results also suggest that sRNAs derived from other PLMVd regions may downregulate additional peach

genes, ultimately resulting in other symptoms or in a more favorable host environment for viroid infection.

Keywords: chloroplast development, heat-shock protein 90, non-coding RNAs, PLMVd, RNA silencing, viroid

pathogenesis.

INTRODUCTION

Viroids are minimal RNAs that infect and often cause severe

diseases in plants (Flores et al., 2005; Tsagris et al., 2008;

Ding, 2009). Based on their properties, among which the

ability to replicate through specific pathways of a rolling-

circle mechanism in certain subcellular compartments is

key, viroid species are classified into two families. The family

Pospiviroidae, type species Potato spindle tuber viroid

(PSTVd) (Diener, 1972; Gross et al., 1978), clusters viroids

replicating and accumulating in the nucleus, whereas the

family Avsunviroidae, type species Avocado sunblotch

viroid (ASBVd) (Symons, 1981; Hutchins et al., 1986), includes

viroids with hammerhead ribozymes in both polarity strands

that mediate self-cleavage of their replicative intermediates

generated in plastids (mostly chloroplasts) (Flores et al.,

2000). Despite being composed of just a small (246–401 nt),

circular, non-protein-coding RNA, viroids can usurp and

redirect the host machinery for completing their infectious

cycle. Therefore, viroids largely differ from viruses, the

replication, movement and pathogenesis of which partly

rely on proteins encoded in their own genome. However,

similarly to changes incited by viruses, viroid infections

cause profound changes in their host homeostasis (Itaya

ª 2012 The Authors 991The Plant Journal ª 2012 Blackwell Publishing Ltd

The Plant Journal (2012) 70, 991–1003 doi: 10.1111/j.1365-313X.2012.04940.x

et al., 2002; Tessitori et al., 2007; Wang et al., 2011), which

ultimately result in the onset of visible symptoms.

RNA silencing, a regulatory network that modulates host

gene expression and protects plants and most other eukar-

yotes against invading nucleic acids, such as transposons,

viruses and transgenes, is triggered by double-stranded

RNAs (dsRNAs) and highly-structured single-stranded RNAs

(ssRNAs) that are processed by Dicer-like (DCL) RNases into

small RNAs (sRNAs) (Carthew and Sontheimer, 2009; Chen,

2009). The two major classes of sRNAs are microRNAs

(miRNAs) and small interfering RNAs (siRNAs), with host

RNA-dependent RNA polymerases (RDRs) mediating an

amplification circuit resulting into secondary siRNAs able

to promote RNA silencing in a non-cell-autonomous manner

(Voinnet, 2008; Dunoyer and Voinnet, 2009). The sRNAs are

loaded into AGO proteins and guide the RNA-inducing

silencing complex (RISC) to target their cognate RNAs or

DNAs (Mallory and Vaucheret, 2010).

To defend themselves, plant viruses encode in their

genomes proteins suppressing the RNA-silencing response

that they trigger in their hosts (Csorba et al., 2009; Ding,

2010). As a side effect, these RNA-silencing suppressors

(RSSs) may impair host developmental pathways regulated

by RNA silencing, explaining at least in part the phenotypic

alterations accompanying most virus infections (Kasschau

et al., 2003; Jay et al., 2011). However, in the absence of

viroid-encoded RSSs, this model cannot account for symp-

toms induced by viroids. In addition to the conventional

model regarding a direct interaction of the genomic viroid

RNA with host proteins and/or RNAs as the primary event of

viroid pathogenesis (reviewed by Flores et al., 2005), an

intriguing alternative model was first proposed when viroid-

derived sRNAs (vd-sRNAs), structurally similar to sRNAs,

were detected in PSTVd-infected tissues (Papaefthimiou

et al., 2001). Accordingly, symptoms would result from

vd-sRNAs loading RISC and targeting specific host mRNAs

for inactivation. This hypothesis has been extended to

explain pathogenesis of satellite RNAs of plant viruses

(Wang et al., 2004) and to involve RDR6-derived secondary

siRNAs in viroid pathogenesis (Gomez et al., 2008).

Recently, direct experimental evidence has been supplied

showing that the yellow symptoms induced in tobacco by

the concurrent infection of Cucumber mosaic virus (CMV)

and its Y-satellite RNA (Y-sat) result from silencing the

chlorophyll biosynthetic gene CHLI by a 22-nt Ysat-derived

siRNA (Shimura et al., 2011; Smith et al., 2011). However, in

spite of multiple reports on the identification and character-

ization of vd-sRNAs accumulating in tissue infected by

nucleus- and chloroplast-replicating viroids (Papaefthimiou

et al., 2001; Martınez de Alba et al., 2002; Itaya et al., 2007;

Di Serio et al., 2009, 2010; Navarro et al., 2009; St-Pierre

et al., 2009; Bolduc et al., 2010; Martınez et al., 2010),

attempts to validate this hypothesis for viroids have failed

so far, with the evidence being circumstantial and restricted

to correlating symptom severity with vd-sRNA accumulation

(Markarian et al., 2004; Wang et al., 2004, 2011; Matousek

et al., 2007; Gomez et al., 2008; Diermann et al., 2010).

Moreover, in most instances a similar correlation also exists

with the genomic viroid RNA, making it difficult to draw

reliable inferences (Itaya et al., 2001; Carbonell et al., 2008;

Di Serio et al., 2010).

To address the issue of whether vd-sRNAs have a role in

pathogenesis, Peach latent mosaic viroid (PLMVd) (Hernan-

dez and Flores, 1992) seems particularly convenient. PLMVd,

a member of the family Avsunviroidae that replicates in the

chloroplast (Bussiere et al., 1999; Rodio et al., 2007), may

induce in its natural host (Prunus persica, peach) a broad

variety of symptoms that includes a severe albinism (peach

calico, PC) in leaves, stems and fruits. PC is associated with

variants of 348–351 nt containing a specific insertion of

12–14 nt (Malfitano et al., 2003) that is absent in latent and

mosaic-inducing variants of 335–338 nt (Ambros et al., 1998,

1999; Flores et al., 2006).

Interestingly, the pathogenic determinant for PC has been

mapped at this insertion, which folds into a hairpin capped

by a UUUU loop (Malfitano et al., 2003; Rodio et al., 2006).

Further dissection of the molecular pathway underlying PC

has shown that PLMVd variants inducing this syndrome

interfere with the maturation of plastid rRNAs, thus impair-

ing plastid translation and chloroplast biogenesis (Rodio

et al., 2007). The recent release of the complete sequence of

the peach genome (peach v1.0; International Peach Genome

Initiative, http://www.rosaceae.org/peach/genome), pro-

vides a critical resource for genome-wide analyses of

PLMVd-induced alterations in this host. Here, we report that

the sequence of the pathogenic determinant for PC is crucial

for eliciting this symptomatology, and supply direct evi-

dence supporting that vd-sRNAs containing the PC deter-

minant program RISC for cleaving a host mRNA, a finding

with potential implications on viroid pathogenesis.

RESULTS

Molecular dissection of the pathogenic determinant for PC

In a previous work we showed that, in addition to the cap-

ping UUUU loop, the stem of the hairpin insertion seems to

play a role in PC (Figure 1a) (Rodio et al., 2006). To further

investigate this role, we selected two natural PLMVd variants

with the same capping UUUU loop but differing stems: the

reference PC-inducing variant C40 with a 4-bp stem (Malfit-

ano et al., 2003) (Figure 1b, left), and variant P1.148 with a

3-bp stem that is unable to elicit this syndrome (Rodio et al.,

2006) (Figure 1c, left). Next, the C and A at position 349 in

variants C40 and P1.148, respectively, were changed into A

and C to generate the artificial variants C40(A349) and

P1.148(C349), in which the predicted stem of the hairpin

insertion was shorter and longer than in their respective

wild-type counterparts (Figure 1b–e, left panels). Slash

992 Beatriz Navarro et al.

ª 2012 The AuthorsThe Plant Journal ª 2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 991–1003

inoculation of dimeric transcripts from the natural and

mutated C40 and P1.148 variants, and dot-blot hybridization

of leaf RNA preparations, showed that the four variants were

highly infectious (the eight GF-305 peach seedlings of each

block became infected) and generated similar viroid titers

(data not shown). These plants were observed for symptom

expression and, to establish a relationship between the

structure and pathogenicity of the infecting variants, their

progeny were cloned and sequenced 6 months post inocu-

lation (6 mpi). To better define this relationship, the albino

sectors of symptomatic leaves were carefully separated

from the adjacent green tissues before extraction, as in a

previous study (Rodio et al., 2006). In addition, for getting a

more complete view of variant distribution within the

PC-expressing seedlings, leaves from an asymptomatic

(green) branch were also analyzed.

As anticipated, variant C40 elicited severe PC symptoms in

the eight inoculated seedlings, whereas variant P1.148 did

not cause any leaf symptom (Figure 1b,c). In line with earlier

results (Malfitano et al., 2003; Rodio et al., 2006), most C40

progeny variants accumulating in albino tissues presented a

minor modification (a G fi A substitution, not disrupting the

hairpin stem) (Figure 1b), whereas in two of the eight

sequenced variants additional minor modifications (includ-

ing a covariation and a substitution) enlarged the stem up to

5 bp (Figure 1b). Intriguingly, all PLMVd variants recovered

(a)

(b)

(c)

(d)

(e)

Figure 1. Peach phenotype and progeny of nat-

ural and artificial PLMVd variants with specific

hairpin insertions.

(a) Partial and complete albinism induced by

variant C40 (left), and schematic representation

of the secondary structure of PLMVd, with the

hairpin insertion represented with a broken line

(right).

(b–e) Results of bioassays with the natural vari-

ants C40, P1.148 (b, c) and their mutant deriva-

tives C40(A349) and P1.148(C349) (d, e), the

hairpin insertion of which (with light-blue back-

ground) plus the flanking 5¢ and 3¢ nucleotides

are depicted on the left. Nucleotides in blue (d, e,

left) denote changes introduced by site-directed

mutagenesis at position 349 in variants C40 and

P1.148 (b, c, left), respectively. Nucleotides in red

denote recurring substitutions emerging natu-

rally with respect to the parental hairpin inser-

tions, with additional, less frequent substitutions

indicated with circles, together with the variants

in which they were found. The ratio of symp-

tomatic to infected seedlings is reported on the

top of each panel in brackets. The sequence of

the hairpin insertions and the name of the

progeny variants recovered from albino (a) and/

or green (g) tissues of one representative inoc-

ulated seedling are also indicated (b–e, middle

and/or left).

Viroid-induced RNA silencing of host mRNA 993

ª 2012 The AuthorsThe Plant Journal ª 2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 991–1003

from the asymptomatic branch of the same plant contained

a similar extended 5-bp hairpin stem (Figure 1b, right panel).

On the other hand, half of P1.148 progeny variants preserved

the hairpin folding of the parental insertion, but the other

half showed longer or shorter stems (Figure 1c).

Unexpectedly, all seedlings inoculated with the mutant

variant C40(A349), and six of eight seedlings inoculated with

the mutant variant P1.148(C349), also displayed PC (Fig-

ure 1d,e). However, the two parental variants were not

recovered in their respective progeny. Instead, in both cases,

the hairpin insertions of variants accumulating in albino and

green tissues were similar, respectively, to those of variants

from albino and green tissues of seedlings inoculated with

their corresponding natural variants (Figure 1, compare

panels d and e with b and c).

Altogether, results from these bioassays show that the

typical hairpin insertion (composed of a UUUU loop and a

4-bp stem previously associated with the pathogenic deter-

minant for PC) was prevalent in PLMVd progeny variants

from albino tissues, irrespective of the specific hairpin

insertion of the inoculated natural or mutant variant. In

contrast, and also irrespective of the specific hairpin inser-

tion of the inoculated variant, PLMVd progeny from green

tissues mostly presented a hairpin insertion with a UUUU

loop but with a 5-bp stem resulting from an additional A:U

pair. These data strongly suggested that the latter variants

did not incite PC. To provide direct support for this hypoth-

esis, we inoculated eight GF-305 peach seedlings with a

dimeric transcript of variant C40(A349)-g12, recovered from

the progeny of variant C40(A349) in the previous experi-

ment, and with an additional A:U pair in the 5-bp stem

(Figure 1d, right panel). All inoculated seedlings remained

asymptomatic and most progeny variants presented the

same hairpin insertion (Figure S1), thereby showing that

variants of this kind cannot elicit PC.

Collectively, these data support the view that the struc-

tural requirements conferring PC depend not only on the

capping loop (UUUU) of the hairpin insertion, but also on

the stem composition (Rodio et al., 2006). However, at this

stage we could not clearly discern whether the sequence or

the specific morphology of the hairpin insertion were

required for inducing PC.

The primary rather than the secondary structure of the

pathogenic determinant plays a major role in PC

We showed previously that the insertion found in some

PLMVd variants always adopts a hairpin folding (Malfitano

et al., 2003; Rodio et al., 2006). To examine whether the

sequence of this hairpin also plays a role in PC, we generated

the artificial variant C40(stem), which differs from its

parental C40 in having the eight nucleotides forming the

stem interchanged (Figure 2). Therefore, the sequence of the

12-nt hairpin insertion of variants C40(stem) and C40 are

largely different, whereas their secondary structure is pre-

served.

Then, dimeric transcripts of these two variants were slash-

inoculated in two blocks of eight peach seedlings each. Dot-

blot hybridization showed that all inoculated plants were

infected at 6 mpi, and that their viroid titer was similar (data

not shown). Remarkably, plants inoculated with variant

C40(stem) remained asymptomatic, with all progeny vari-

ants preserving the stem stability and some displaying

mutations in the capping loop (Figure 2). In contrast, control

seedlings inoculated with variant C40 expressed typical PC

symptoms. Therefore, these data show that preservation of

the secondary structure (folding into a hairpin with a

capping UUUU loop) is not sufficient for conferring PC

pathogenicity to this structural element, and strongly sup-

port that its primary structure, particularly the 12-nt

sequence GA(A/G)CUUUUGUUC (hereafter named PC-asso-

ciated insertion) characteristic of variants from albino

tissues, is critical in this respect.

PLMVd-sRNAs derived from the pathogenic determinant for

PC accumulate during infection

In view of the apparent role played by the sequence of the

PC-associated insertion, we examined whether pathogenesis

could operate by the silencing-based model proposed pre-

viously for viroids and plant satellite RNAs (Wang et al.,

2004). Adapting this model to our system implicitly entails

that (+) or ()) PLMVd-sRNAs involved in PC should contain,

at least partially, the PC-associated insertion.

To explore this possibility we subjected the sRNAs

accumulating in PC-expressing seedlings to deep sequencing.

Inoculated Progeny in green tissue

C40(stem)-g03, C40(stem)-g09, C40(stem)-g11

C40(stem)

U U U

U G C .

A U .

G C .

A C

G . U

C40(stem) (symptomatic/infected: 0/8)

U C U

U G C .

A U .

G C .

U C

G . C

Ag11 A g11

U g09 U

U U U

G C .

A U .

G C .

A C

A . U

A g07 g08

U C U A

G C .

A U .

A U .

A C

G . C

C40(stem)-g06, C40(stem)-g07, C40(stem)-g08

U U U

U G C .

A U .

A U .

A C

G . C

C40(stem)-g02 C40(stem)-g12

A-

Figure 2. Peach phenotype and progeny of the artificial PLMVd variant C40(stem).

Nucleotides in blue (left) denote changes introduced by site-directed mutagenesis in the hairpin insertion of natural variant C40 to interchange the nucleotides of the

hairpin stem, preserving its stability. Other details are as described in the legend to Figure 1.

994 Beatriz Navarro et al.

ª 2012 The AuthorsThe Plant Journal ª 2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 991–1003

This approach has already been adopted in a study aimed at

comparing the vd-sRNAs accumulating in leaves of GF-305

peach seedlings infected with the PC-inducing variant C40

and a mosaic-inducing PLMVd variant (GDS6) (Di Serio

et al., 2009). This previous C40 sRNAs library was prepared

from symptomatic leaves that also contained green sectors.

Considering the role of the PC-associated insertion in

pathogenesis and the uneven distribution of variants with

this structural element in symptomatic plants (see above),

we generated a C40 sRNA library from albino leaf tissues

carefully separated from the surrounding green sectors in

order to enrich it with vd-sRNAs from variants containing the

PC-associated insertion. In parallel, we also generated an

sRNAs library from leaves of a mock-inoculated peach

seedling as a negative control. Gel-purified sRNAs were

linked to bar-coded adaptors for deep sequencing of both

libraries in the same channel.

Of approximately 20 815 300 reads, 96.2% were attribu-

table to the two bar-coded sRNA preparations (53.2 and

46.8% for the C40 and the mock-inoculated sample, respec-

tively). Reads for sRNAs between 18 and 26 nt (9 014 628

and 7 557 838 for the C40 and the mock-inoculated sample,

respectively), adopted a profile with two prominent 21- and

24-nt peaks in both cases (Figure S2). When PLMVd-sRNAs

matching perfectly the (+) and ()) sequences of variant C40

and its progeny (Figure 1b) were searched, 1 925 916 reads

(21.3% of the total generated by sRNAs between 18 and

26 nt) were retrieved from the C40 sample, corresponding to

8015 non-redundant sequences (Appendix S1), with minimal

PLMVd-related sequences detected in the mock-inoculated

sample (3961 reads representing 0.05% of the total, 611 of

which were non-redundant); the latter reads most likely

result from a minor contamination, because none of these

PLMVd-related sequences matched perfectly with the peach

genome. In line with results of the previous deep-sequenc-

ing experiment (Di Serio et al., 2009), (+) PLMVd-sRNAs

(57.6%) were slightly more abundant than their ()) counter-

parts (42.4%), with a similar polarity distribution being

observed after segregating PLMVd-sRNAs in size classes

(Figure S3). Moreover, also in line with previous results (Di

Serio et al., 2009), size classes of 21 and 22 nt were largely

prevalent (64.5 and 23.8%, respectively) within the PLMVd-

sRNAs population, whereas a minor fraction of 20-, 23- and

24-nt vd-sRNAs (around 10.2% altogether) was retrieved

(Figure S3). Mapping the PLMVds-RNA reads along the

genomic (+) and ()) RNAs revealed specific hot-spot profiles

(Figure S4). A parallel mapping of the non-redundant

PLMVd-sRNAs also revealed profiles with hot spots corre-

sponding, as expected, to genomic regions with high

nucleotide variability in the progeny variants (Figure S5).

A search for PLMVd-sRNAs spanning totally or partially (in

at least one nucleotide) the sequence of the PC-associated

insertion from C40 and its progeny variants revealed 111

non-redundant (+) and ()) 21-nt RNAs (6993 reads). Addi-

tionally, 100 non-redundant (+) and ()) vd-sRNAs of 22 nt

(6799 reads) were mapped at the same structural element.

These data show that the PC-associated insertion is indeed a

source of vd-sRNAs (hereafter referred to as PC-sRNAs),

although at this stage we did not know whether, like genuine

miRNAs, they might target host mRNAs for inactivation.

Host mRNAs are potentially targeted by PC-sRNAs

To search transcripts that could be targeted by PLMVd-

sRNAs for RISC-mediated degradation, we took advantage

of the complete sequence of the peach genome released

recently (peach v1.0; International Peach Genome Initiative,

http://www.rosaceae.org/peach/genome), and applied the

RNAhybrid program (Rehmsmeier et al., 2004). The search

was restricted to sRNAs of 21 nt because this is the prevalent

size of PLMVd-sRNAs and plant miRNAs (Di Serio et al.,

2009; St-Pierre et al., 2009; Figure S3). Moreover, we took

into consideration that miRNA-directed mRNA cleavage in

plants requires a high degree of base pairing, and that

mismatches in bona fide miRNA:target duplexes are mostly

located in the first position and after position 13 relative to

the 5¢ terminus of the miRNA (Mallory et al., 2004). There-

fore, a position-dependent scoring matrix for predicting

plant miRNA targets (Fahlgren and Carrington, 2010)

was applied to all possible duplexes formed by 21-nt

PLMVd-sRNAs and predicted peach mRNAs identified as

potential targets. A maximum score cut-off of 2.5, which is

below the cut-off of 3.5 established for reliable sensitivity

and specificity in Arabidopsis thaliana (Fahlgren and

Carrington, 2010), was adopted. This cut-off of 2.5 seemed

appropriate for our search because the duplexes between

miRNAs identified in our libraries conserved in other species

(including miR156, miR159, miR160, miR162, miR164,

miR166, miR168, miR171, miR172, miR393, miR394, miR395,

miR396 and miR397) and their potential targets in the com-

plete peach genome showed scores from 0 to 3. The search

resulted in 65 predicted peach mRNA targets, most encoding

proteins with functional annotation (P. persica – JGI v1.0;

http://www.phytozome.net/cgi-bin/gbrowse/peach; Table

S1), and 30 of them having been detected as peach tran-

scripts (peach assembled expressed sequence tags, ESTs,

from peachv1pasa_assemblies).

Considering that the sequence of the PC-associated

insertion is critical for eliciting PC, and that this insertion is

a source of vd-sRNAs (PC-sRNAs) in albino tissues (see

above), we next concentrated on the duplexes formed by

PC-sRNAs and peach mRNAs detected as transcripts

(Table 1). Overall, 12 PC-sRNAs were found potentially

targeting 10 predicted peach mRNAs because PC-sRNA1

and PC-sRNA6 could target two different mRNAs (Table 1).

This table also shows that two of the ten encoded proteins

contain a predicted chloroplastic transit peptide (cTP) (Li and

Chiu, 2010), according to CHLOROP software (http://www.

cbs.dtu.dk/services/ChloroP) (Emanuelsson et al., 1999),

Viroid-induced RNA silencing of host mRNA 995

ª 2012 The AuthorsThe Plant Journal ª 2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 991–1003

thus indicating that PC-sRNAs might impair the expression

of nuclear genes coding for plastid proteins. Nine and three

PC-sRNAs were of (+) and ()) polarity, respectively (Table 1).

Altogether, these data show that PC-sRNAs potentially

targeting peach mRNAs are actually generated in albino

tissues, opening the possibility that at least some could act

like endogenous miRNAs or siRNAs.

The mRNA encoding the chloroplastic heat shock protein 90

(cHSP90) is targeted by PC-sRNAs for sequence-specific

degradation

To provide direct evidence for the sequence-specific degra-

dation of a peach mRNA mediated by PC-sRNAs, we focused

on peach transcript ppa001590m for several reasons. First,

two PC-sRNAs (PC-sRNA8a and PC-sRNA8b) were identified

potentially targeting this transcript for cleavage, with the

corresponding duplexes showing scores ranging among the

lowest recorded (Table 1). Second, PC-sRNA8a and

PC-sRNA8b, which derive from the viroid ()) strand, completely

span the PC-associated insertion. And third, ppa001590m

transcript codes for a protein of 797 amino acids annotated

as homologous with CR88 of A. thaliana (84.8% similarity

with AT2G04030.1), a chloroplast-targeted HSP90 protein

(cHSP90) (Cao et al., 2003). A multiple alignment using

CLUSTALW (Thompson et al., 1994) showed that peach

cHSP90 also displays high similarity with cHSP90 proteins

from other plant species and Chlamydomonas reinhardtii

(Figure 3), including the four domains conserved in the

HSP90 family (Meyer et al., 2003), and the DPW C-terminal

motif exclusively found in the plastidic members (Chen

et al., 2006). Moreover, within the N-terminal 67 amino acids

of peach cHSP90, the CHLOROP software predicted a cTP with

a length and score similar to those obtained for the homol-

ogous proteins from the other species (Table S2). Therefore,

these data strongly suggest that peach cHSP90 is a nuclear-

encoded protein with plastidic localization. Interestingly,

PC-sRNA8a and PC-sRNA8b target the sequence coding for

the cTP (Figure 3). Furthermore, peach cHSP90 appears

Table 1 Predicted peach mRNAs annotated as transcripts and targeted by PLMVd-sRNAs derived from the PC-associated insertion

aPC-sRNAs sequenced from an sRNA library from albino leaf sectors resulting from infection with PLMVd variant C40; following the criterionadopted for miRNAs, PC-sRNAs differing in one position are denoted with letters a–c, following the serial number.bLocation of PC-sRNA 5¢ termini in the genomic RNAs of variant C40.cmRNA:PC-sRNA duplexes predicted by RNA hybrid (Rehmsmeier et al., 2004); targeted host mRNAs are on top (5¢ fi 3¢ orientation) and PC-sRNAsbelow (3¢ fi 5¢ orientation); nucleotides spanning the PC-associated insertion are in bold; |, Watson–Crick base pairs; o, G:U wobble base pairs; ),mismatches or buldges.dMinimum free energy (mfe, kcal mol)1) of the mRNA target:PC-sRNA duplex.eScore of the mRNA target:PC-sRNA duplex, estimated as in Fahlgren and Carrington (2010); the lower the score, the more reliable the prediction.fAnnotations from Prunus persica – JGI v1.0 (http://www.phytozome.net/cgi-bin/gbrowse/peach).gChloroplast targeting of the protein predicted by CHLOROP software (http://www.cbs.dtu.dk/services/ChloroP).

996 Beatriz Navarro et al.

ª 2012 The AuthorsThe Plant Journal ª 2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 991–1003

particularly appealing in the context of PC pathogenesis

because its homologous CR88 is required for chloroplast

biogenesis in A. thaliana (Cao et al., 2003); this protein,

perhaps interacting with HSP70, may be needed for assem-

bling the core of a multichaperone complex involved in the

biogenesis/maintenance of thylakoid membranes (Schroda

and Muhlhaus, 2009), which is the developmental pathway

specifically altered in PC-expressing tissues (Rodio et al., 2007).

We next proceeded comparing the steady-state level of

cHSP90 mRNA in mock-inoculated and PLMVd-infected

peach seedlings expressing PC. In line with data from the

peach EST database, RT-PCR amplification of a 405-bp cDNA

corresponding to the 5¢ terminus of the peach mRNA for

cHSP90, followed by cloning and sequencing, showed that

the gene cHSP90 is indeed transcribed in the mock-inocu-

lated control. However, RT-PCR estimates using the same

primers and serial cDNA dilutions showed a substantially

lower steady-state level of the cHSP90 transcript in albino

leaf sectors than in green tissues [from the same

C40-inoculated seedling, or from a mock-inoculated seed-

ling or a seedling infected by the latent variant C40(A349)-

g12]. As additional controls we also tested the levels of the

transcripts from genes psbA and rpoB, which were lower

and higher in albino than in green tissues, respectively

(Figure 4), in agreement with a previous report showing that

mRNAs from genes transcribed by the plastid-encoded

polymerase (like psbA) are negatively affected in albino

tissues, whereas those transcribed by the nuclear-encoded

polymerase (like rpoB) over-accumulate (Rodio et al., 2007).

The lower accumulation of the cHSP90 transcript in albino

tissues is consistent with its targeting by PC-sRNA8a and

PC-sRNA8b for RISC-mediated degradation (Figure 5). This

hypothesis was validated by RNA ligase-mediated rapid

amplification of cDNA ends (RLM-RACE) experiments with

RNA preparations from albino tissues: four out of five cDNA

clones were from fragments of the cHSP90 transcript with

5¢ termini identical to the cleavage site predicted by

PC-sRNA8a and PC-sRNA8b (Figure 5), between positions

10 and 11 from their 5¢ termini, with the cleavage site

inferred from the other clone mapping at one nucleotide

| Nucleotide binding domainA. thaliana -MAPALSRSLYTSPLTSVPITP-----VSSRLSHLRSSFLPHGGALRTGVSCS---WNLEK--RCNRFAVKCDAAVAEK--ETTEEGSGEKFEYQAEVSRLLDLIVHSLYSHKEVFLREL 107P. persica -MAPVLSRSLATASLASLPSSSPFTLRNPSKALSLRSAFVPQNG-LRKGFSCGGLKWKLES--KNRGISIRCDAAVAEK--EATDT-PGEKFEYQAEVTRLMDLIVHSLYSHKEVFLREL 113S. cereale -MAPALSRTLGPSSVAALRPSP-------SRGLPLAALLPQGKRSS-SARGVR---WEAG---RGRLVGARCASAVAEKTAGEEEEAAGEKFEYQAEVSRLMDLIVHSLYSHKEVFLREL 105O. sativa -MAPSLSRSLGASSVAALRPCA-------GRVRAPGAGAARGSGSARCGRGVR---WEAGSGSRGRLVRVRCDAAVAEK---AEETAEEEKFEYQAEVSRLMDLIVHSLYSHKEVFLREL 106C. reinhardtii MMLRGLSSGMRQARAQSSASAAS-----AARPLPLLARGNGVSSALLSGASPC----AAVTAAASLRPLPAGRGPVLMRAAATEAASGSETFTYQAEVDRLMDMIVNSLYSNREVFLREL 111

ATP binding sitenucleotide binding domain

A. thaliana VSNASDALDKLRFLSVTEPSLLGDGGDLEIRIKPDPDNGTITITDTGIGMTKEELIDCLGTIAQSGTSKFLKALKENKDLGADNGLIGQFGVGFYSAFLVAEKVVVSTKSPKSDKQYVWE 227P. persica VSNASDALDKLRFLSVTEPSLLGDAGELQIRIKPDPDNGTITITDTGIGMTKEELIDCLGTIAQSGTSKFLKALKENKDLGADNGLIGQFGVGFYSAFLVAEKVVVSTKSPRSDKQYVWE 233S. cereale VSNASDALDKLRFLSVTDSSVLADGGELEIRIKPDPDAGTITITDSGIGMTKDELKDCLGTIAQSGTSKFLKALKENKELGADNGLIGQFGVGFYSAFLVAEKVVVSTKSPKTDKQYIWE 225O. sativa VSNASDALDKLRFLGVTDSSLLADGGELEIRIKPDPDAGTITITDTGIGMTKDELKDCLGTIAQSGTSKFLKALKENKDLGADNGLIGQFGVGFYSAFLVAEKVVVSTKSPKSDKQYVWE 226C. reinhardtii ISNASDALDKARFLSLTDPSVLAGREELDIRISADKEKGTLVIEDSGIGMSREQLLSNLGTIARSGTRKFMEAMAAKG----DTNLIGQFGVGFYSAFLVADRVMVQSKSPEEAKHWVWE 227

Nucleotide binding domain | Charged linke | Middle domain A. thaliana SVADSSSYLIREETDPDNILRRGTQITLYLREDDKYEFAESTRIKNLVKNYSQFVGFPIYTWQ-----------EKSRTIEVEEDEPVKEGEE-GEPK--KKKTTKTEKYWDWELANETK 333P. persica AAADSSSYVIREETDPENLIRRGTQITLYLRPDDKYEFSEPARIQGLVKNYSQFVSFPIYTWQ-----------EKSRTVEVEEEEEPKEGEE-PKPEGEKKKKTKTEKYWDWELANETK 341S. cereale AEANSSSYVIREETDPEKMLTRGTQITLFLREDDKYEFADPARIQGLVKNYSQFVSFPIFTWQ-----------EKSRTVEVE-EEESKEGEETAEGEKEEKKKTITEKYWDWELANETK 333O. sativa GVADSSSYVIKEETDPEKMLTRGTQITLVLRPDDKFEFADPGRIQGLVKNYSQFVSFPIYTWQ-----------EKSRTVEVEEDEEAKEGEEAKEGE-QKKKKTITEKYWDWELANETK 334C. reinhardtii AKAGSHQYSIREDEAKD--LVRGTRITLYLK-EDAAEMADTVKITQLIKQYSQFIAFPIKVYAPKKEPRKVVDEEATKKKQAAADAKAKEAGEEAAKPVEPVMKTEYDEVWDWRLENENK 344

Middle domainA. thaliana PLWMRNSKEVEKGEYNEFYKKAFNEFLDPLAHTHFTTEGEVEFRSILYIPGMGPLNNEDVTNPKTKNIRLYVKRVFISDDFDGELFPRYLSFVKGVVDSDDLPLNVSREILQESRIVRIM 453P. persica PIWMRNPKEVEKDEYHEFYKKTFSEFLDPVAYTHFTTEGEVEFRSVLYIPGMGPLNNEDVVNAKTKNIRLYVKRIFISDDFDGELFPRYLSFVKGVVDSNDLPLNVSREILQESRIVRIM 461S. cereale PIWMRNPKEVEETEYNEFYKKAFNEFLDPLAHAHFTTEGEVEFRSVLYIPGMAPLSNEEIMNPKTKNIRLYVKRVFISDDFDGELFPRYLSFVKGVVDSNDLPLNVSREILQESRIVRIM 453O. sativa PIWMRSPKEIEKTEYNEFYKKAFNEFLDPLAYTHFTTEGEVEFRSVLYIPGMAPLSNEEIMNPKTKNIRLYVKRVFISDDFDGELFPRYLSFVKGVVDSNDLPLNVSREILQESRIVRIM 454C. reinhardtii PIWTRSPKDVSETAYNDFFKTTFGEFLDPLAHVHFNVEGTIEFSSILYIPGMAPFEQQNMQ-QRSKSIKLYVKRVFISDEFDEDLMPRYLAFVKGVVDSSDLPLNVSREILQESRIVRVI 463

Middle domainA. thaliana RKRLIRKTFDMIQEISESENKEDYKKFWENFGRFLKLGCIEDTGNHKRITPLLRFFSSKNEEELTSLDDYIENMGENQKAIYYLATDSLKSAKSAPFLEKLIQKDIEVLYLVEPIDEVAI 573P. persica RKRLVRKTFDMIQEISESENKEDYKKLWENFGRFLKLGCIEDSGNHKRLTPLLRFYSSKSEEELISLDDYVENMPENQKAIYYLAADSLKSAKSAPFLEKLVQKDIEVLYLVEPIDEVAI 581S. cereale RKRLVRKTFDMIQDIADKDNKEDYKKFGESFGKFMKLGCIEDSGNQKRLAPLLRFYSSKNETDLISLDQYVENMPETQKAIYYIATDSLQSAKTAPFLEKLLQKDIEVLYLIEPIDEVAI 573O. sativa RKRLVRKTFDMIEEISEKDDKEDYKKFWESFGKFIKLGCIEDTGNHKRLAPLLRFHTSKNEGDLISLDQYVENMPENQKAIYYIATDSLQSAKTAPFLEKLVQKDIEVLYLIEPIDEVAI 574C. reinhardtii RKQLVRRSIEMLEELAGKEGGEDYKTFWEAFGRNIKYGVIEDTENRERLSKLLRFSSSKAEDSLTSLDEYVGRMGANQKTIYYMAADSVAAARAAPFMEAMVAKGIEVLYLTEPIDEACV 583

Middle domain | Dimerization/co-chaperoneA. thaliana QNLQTYKEKK------FVDISKE--DLELGDEDEVKDREAKQEFNLLCDWIKQQLGDKVAKVQVSNRLSSSPCVLVSGKFGWSANMERLMKAQALGDTSSLEFMRGRRILEINPDHPIIK 685P. persica QNLQTYKEKK------FVDISKE--DLELGDEDEVKERETEQEYNLLCDWIKQQLGDKVAKVQVSKRLSSSPCVLVSGKFGWSANMERLMKAQALGDTASLEFMRGRRILEVNPDHPIVK 693S. cereale QNLQTYKEKK------FVDISKE--DLELGDEDEDKE-ETKQEYTLLCDWIKQQLGDKVAKVQISKRLSSSPCVLVSGKFGWSANMERLMKAQTLGDTSSLEFMRGRRIFEINPDHPIVK 684O. sativa QNLQTYKEKK------FVDISKE--DLELGDEDEDNK-ESKQEYTLLCDWVKQQLGDKVAKVQISKRLSLSPCVLVSGKFGWSANMERLMKAQTLGDTSSLEFMRGRRIFEINPDHPIVK 685C. reinhardtii TNLGKYGPDKNGPQYELVDVSKEGVSLDEGEDEKKKAEEVAKDMAPVVDFLKKALGERVEKVTVSNRLLDSPCALVTSKFGWSANMERIMRSQALGDARAMEYMKGRKIMEINPNHDIIA 703

interaction Dimerization/co-chaperone interaction |A. thaliana DLNAACKNAPESTEATRVVDLLYDTAIISSGFTPDSPAELGNKIYEMMAMAVGGRWGRVEEEEESSTVNEGDDKSG---------ETEVVEPSEVRAES---DPWQD-- 780P. persica DLKAACKNAPDSTDAKRAVDLLYDTALISSGFTPDSPAELGNKIYEMMAMALGGRWGRSEDEEAETEVADGDAGESSDASAGEVVEAQVIEPSDVRAES---DPWSD-- 797S. cereale DLSAACKNEPDSTEAKRAVELLYETALISSGYTPESPAELGGKIYEMMTIALGGRWGRS-----GAEEAETNVDGDSS----EGVVPEVIEPSEVRTENE-NDPWRD-- 781O. sativa DLNAACKNEPESTEAKRAVELLYETALISSGYTPDSPAELGGKIYEMMTIALGGRWGRSDTETEAATTGDASTETGSS----EATVTEVIEPSEVRPES---DPWRD-- 785C. reinhardtii GIKTLLK-EKDEDRARDLSELLYETALITSGFQVDSPKDYASKVFTLMKIALG-YDILSEAEEQAAAAAPQAAEAAAAPKAAEAAAVPKVEATPVDAEVVSDDPWKKSA 810

Figure 3. Comparison of chloroplast-targeted cHSP90 from different species.

Alignment of the amino acid sequences of cHSP90 from Prunus persica, Arabidopsis thaliana (NP_178487) (Cao et al., 2003), Secale cereale (CAA82945) (Schmitz

et al., 1996), Oryza sativa (NP_001062103) (Chen et al., 2006) and Chlamydomonas reinhardtii (XP_001702984) (Willmund and Schroda, 2005). Residues with a black

background are conserved in all sequences, whereas those in grey are conserved in four of the five sequences. Amino acids with similar chemical properties are

denoted as N/Q, D/E, R/K, S/T, F/Y, A/G and V/I/L/M. Domains in cHSP90s are according to Meyer et al. (2003). The C-terminal amino acid of the chloroplast transit

peptide predicted by the CHLOROP program (Emanuelsson et al., 1999) is in red. The C-terminal DPW motif conserved in all chloroplastic HSP90 species is in green

(boxed). Amino acids in blue denote those encoded by the fragment of the peach cHSP90 mRNA predicted to be targeted by PC-sRNA8a and PC-sRNA8b.

Viroid-induced RNA silencing of host mRNA 997

ª 2012 The AuthorsThe Plant Journal ª 2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 991–1003

downstream. These data conclusively showed that

PC-sRNA8a and PC-sRNA8b downregulate a peach mRNA,

as predicted by an RNA silencing mechanism. Conversely,

the lack of the same RLM-RACE products when using RNA

preparations from green tissues [from the C40-inoculated

seedling, or from a mock-inoculated seedling or a seedling

infected by the latent variant C40(A349)-g12], together with

the similar cHSP90 mRNA levels detected in the same

tissues (Figure 4), indicate that this peach transcript is not

targeted for RISC-mediated degradation in the asymptom-

atic tissues. Finally, these findings are also consistent

with the biased distribution of PLMVd variants observed

in symptomatic seedlings (Figure 1): PCsRNA8a and

PC-sRNA8b are exclusively generated from variants con-

taining the PC-associated insertion, which only accumulate

in albino leaf sectors. To give further credence to this view,

two additional peach sRNA libraries were generated and

sequenced from green leaves of PC-expressing and non-

symptomatic seedlings inoculated with variants C40 and

P1.148, respectively (Figure S6). Although PLMVd-sRNAs

deriving from the hairpin insertion were sequenced in both

libraries, none contained the PC-associated insertion. In

particular, PC-sRNA8a and PC-sRNA8b were not found, and

the PLMVd-sRNAs sequenced that could form duplexes with

cHSP90 mRNA were derived from insertions of variants

characteristically accumulating in green tissues. However,

the corresponding scores were high (from 5 to 7.5; see

Figure S7), further supporting the hypothesis that the sRNAs

resulting from PLMVd variants without the PC-associated

insertion do not prime RISC for cHSP90 mRNA cleavage.

DISCUSSION

In the first part of this study we furthered the characterization

of the PLMVd determinant for PC by dissecting albino and

green sectors of GF-305 peach seedlings inoculated with

natural and artificial variants of this viroid, and subsequently

cloning the resulting progeny. This approach has revealed

the 12-nt sequence GA(A/G)CUUUUGUUC as the insertion

conferring PC to infecting PLMVd variants. In agreement

with previous data (Malfitano et al., 2003; Rodio et al., 2006),

this insertion folds into a hairpin capped by a UUUU loop,

and differs in the stem composition from other hairpin

insertions found in PLMVd variants recovered from green

tissues. Therefore, the primary rather than secondary

structure of the pathogenic determinant is involved in elic-

iting PC, as clearly illustrated by the artificial variant

C40(stem). The infecting ability of this variant probably

results from the eight mutations introduced not affecting the

stem stability of the hairpin insertion (Figure 2), in line with

cDNA dilution 1:2 1:20 1:200

cHSP90

Healthy

C40(349)-g12(green)

C40 (albino)

C40 (green)

18S psbA rpoB

1:2 1:20 1:200 1:2 1:20 1:200 1:2 1:20 1:200

Figure 4. Transcript accumulation in GF-305 leaves.

RT-PCR semiquantitative estimation of two nucleus-encoded transcripts (the

18S rRNA and the mRNA of the chloroplast-targeted cHSP90) and two plastid-

encoded transcripts (the mRNAs of the rpoB subunit and of the psbA

component of photosystem II). The cDNAs, generated by random-primed

reverse transcription of equalized RNA preparations, were serially diluted

before PCR amplification with gene-specific primer pairs. The amplified

products were analyzed by agarose gel electrophoresis and ethidium bromide

staining. Only gel sections containing the amplified cDNAs are presented.

Results are representative of three replicates conducted on two biologically

independent samples.

1/5 4/5

5’-..AGCCUGAGAAGUGCUUUUGUUCCUCAAAAUGGGCUCAGGAAGG..-3’||| |||||||||||||||||

3’-UUCUCGAAAACAAGGAGUUUU-5’PC-sRNA8a

cHSP90 mRNA

336 7

AAAA

AA

G

G U UU

U

UU C G A

UCC

C. . . .

7

336

scaffold_1: 28257983..28265359

28258k 28259k 28260k 28261k 28262k 28263k 28264k

Transcriptppa001590m

Figure 5. Specific cleavage of peach cHSP90 mRNA

mediated by PC-sRNAs.

Upper panel, predicted cHSP90 transcript and loca-

tion of the gene into the peach genome v1.0 (http://

www.phytozome.org). Middle panel, cHSP90

mRNA:PC-sRNA8a duplex (in bold), with the

sequence of PC-sRNA8a corresponding to the

PC-associated insertion underlined (note that they

are of complementary polarity). Arrows mark the

predicted and validated cleavage sites by RLM-

RACE from five independent clones, with fractions

indicating the number of clones producing the same

results. Lower panel, schematic representation of

the secondary structure predicted for PLMVd variant

C40 with nucleotides complementary to PC-sRNA8a

highlighted, and those forming part of the PC-induc-

ing insertion on a black background.

998 Beatriz Navarro et al.

ª 2012 The AuthorsThe Plant Journal ª 2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 991–1003

previous observations with another member of the same

genus, Chrysanthemum chlorotic mottle viroid (CChMVd;

Navarro and Flores, 1997), which also remains viable as long

as the introduced mutations preserve elements of its sec-

ondary and tertiary structure (De la Pena et al., 1999; Gago

et al., 2005).

Altogether, these findings are consistent with the silenc-

ing-based model of pathogenesis proposed for viroid and/or

satellite RNAs (Papaefthimiou et al., 2001; Wang et al.,

2004), recently validated experimentally for the yellow

symptoms induced by a specific satellite RNA (Y-sat) of

CMV (Shimura et al., 2011; Smith et al., 2011). However,

regarding viroids, previous attempts to obtain support for

this model have met with limited success. Although some

reports highlight the positive correlation between symptom

severity and the accumulation of vd-sRNAs in different

plant–viroid combinations (Itaya et al., 2001; Markarian

et al., 2004; Matousek et al., 2007), including transgenic

Solanum lycopersicum (tomato) lines expressing a non-

infectious PSTVd hairpin RNA (Wang et al., 2004), others

have failed to find a similar correlation in some transgenic

lines expressing the same non-infectious PSTVd hairpin

RNA, and apparently accumulating similar vd-sRNA titers

(Schwind et al., 2009). Moreover, although RDR6, an enzyme

mediating synthesis of secondary siRNAs, has been

involved in symptom induction by Hop stunt viroid in

Nicotiana benthamiana (Gomez et al., 2008), symptoms

elicited by PSTVd in the same plant species are RDR6-

independent (Di Serio et al., 2010). Setting aside these

contradictory results, none of the previous studies have

provided direct experimental support for the vd-sRNAs

actually mediating cleavage of host mRNAs; at best, a

correlation between the downregulation of tomato mRNAs

potentially targeted by PSTVd sRNAs and symptom expres-

sion has been found (Wang et al., 2011). Finally, even if

vd-sRNAs may target and downregulate, in a sequence-

specific manner, the overexpression of a reporter gene (Vogt

et al., 2004; Itaya et al., 2007) or the accumulation of viroid

RNA in infected plants (Carbonell et al., 2008), these results

do not prove that a similar mechanism operates against host

genes in natural infections.

In the second part of this study we supply solid experi-

mental evidence showing that the accumulation level of the

mRNA coding for the chloroplast-targeted cHSP90 from

peach is reduced in the albino sectors of PC-expressing

leaves. Our data, obtained by a combination of deep

sequencing, semi-quantitative RT-PCR and RLM-RACE, are

consistent with the sequence-specific cleavage of this mRNA

guided by PLMVd-sRNAs (PC-sRNA8a and PC-sRNA8b)

generated from the PC-associated insertion of some PLMVd

variants, thus illustrating that at least this viroid can indeed

modify its host gene expression through RNA silencing.

Chloroplast developmental defects and malfunctioning of

plastid-to-nucleus signaling reported in PC-expressing

tissues (Rodio et al., 2007) support a role for the down-

regulation of cHSP90 in eliciting the albino phenotype,

because this chloroplast-targeted protein has been involved

in chloroplast biogenesis and plastid-to-nucleus signal

transduction in Arabidopsis and Chlamydomonas (Cao

et al., 2003; Willmund and Schroda, 2005; Willmund et al.,

2008). Additional support for this view comes from the

yellow phenotype observed in a mutant line of A. thaliana

with a single amino acid change in cHSP90 (Cao et al., 2003),

and from two cHSP90 T-DNA null-mutant lines (EMB 1956-1

and EMB 1956-2) of A. thaliana that are embryo defective,

and display white embryos and seeds (Meinke et al., 2008).

Although it has recently been shown that the yellow

symptoms induced in Nicotiana tabacum by the Y-sat of

CMV result from silencing the chlorophyll biosynthetic gene

CHLI with a 22-nt siRNA derived from this satellite RNA

(Shimura et al., 2011; Smith et al., 2011), another more

complex RNA silencing-based model has been offered for

explaining the attenuation of the yellowing symptomatol-

ogy of N. benthamiana induced by a different satellite RNA

of the same virus (Hou et al., 2011), indicating that the

pathogenic mechanism proposed for Y-sat may not be

general. Whether the silencing of gene cHSP90 guided by

PC-sRNA8a and PC-sRNA8b is sufficient for inciting PC is not

known because, at this stage, we cannot rule out the

involvement of other factors. However, PLMVd-sRNAs other

than PC-sRNAs can be excluded as direct players in this

respect, because they lack nucleotides derived from the

insertion strictly associated with PC.

Nine and three PC-sRNAs potentially targeting peach

mRNAs have (+) and ()) polarity, respectively (Table 1), but

both PC-sRNA8a and PC-sRNA8b derive from the ()) polarity

strand. These results indicate that the activity of vd-sRNAs in

promoting host mRNA cleavage may not depend on their

relative abundance. In fact, the PLMVd region encompassing

the PC determinant does not map at a hot spot in the profile

of vd-sRNA reads (Figure S4); neither does the Y-sat region

from which the 22-nt (+) species silencing CHLI mRNA

derives (Shimura et al., 2011; Smith et al., 2011). This finding

is in agreement with results of a previous study in which

viroid dsRNAs, probably produced in the cytoplasm by host

RDR(s), were proposed to serve as substrates for generating

PLMVd-sRNAs (Di Serio et al., 2009). Consistent with this

dsRNA-based origin of most PLMVd-sRNAs, the passenger

strand of PC-sRNA8b was also identified in the sRNA library

(Appendix S1, sequence ID: MAC-18-1596391_21_3_0), but

with a number of reads (only three) significantly lower than

that of its guide counterpart (169 reads; Table 1) mediating

cleavage of the cHSP90 mRNA. Two other points are worthy

of note: (i) PC-sRNA1 and PC-sRNA6 may form duplexes

with more than one peach mRNA, suggesting that a small

viroid region has the potential for silencing multiple host

targets; and (ii) the U prevalence in the four 5¢ terminal

positions of PC-sRNA8a and PC-sRNA8b is remarkable

Viroid-induced RNA silencing of host mRNA 999

ª 2012 The AuthorsThe Plant Journal ª 2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 991–1003

because this trait is also found in the 22-nt siRNA from Y-sat

of CMV silencing CHLI mRNA.

As PLMVd accumulates genetic heterogeneity very rap-

idly (Ambros et al., 1999; Rodio et al., 2006; this work), in line

with the observation that the related CChMVd has the

highest mutation rate reported for any biological entity

(Gago et al., 2009), new PLMVd-sRNAs may be generated

with time, targeting other peach genes and eventually

resulting in the fluctuating phenotype characteristic of most

PLMVd infections (Flores et al., 2006). Consistent with this

view, most PLMVd-sRNAs with potential peach targets

derive from viroid regions displaying high sequence vari-

ability (Figure S5). As previously proposed for viruses

(Moissiard and Voinnet, 2006; Llave, 2010), changes in host

gene expression induced by PLMVd-sRNAs could generate a

more favorable environment for viroid infection. Recipro-

cally, viroid modulation of host gene expression via RNA

silencing might play a role in the molecular evolution of

these minimal infectious agents.

EXPERIMENTAL PROCEDURES

Plant material and growing conditions

Leaf tissues were from GF-305 peach [Prunus persica (L.) Batsch]seedlings slash-inoculated with buffer or dimeric head-to-tail tran-scripts generated in vitro from natural and artificial PLMVd variantsinducing different foliar symptoms or no symptoms (see below). Sixweeks after inoculation, symptom expression was stimulated bychilling the seedlings at 4�C in the darkness for 6–8 weeks, and thentransferring them back to the glasshouse to favor the emergence ofnew flushes.

RNA extraction and dot-blot hybridization

Total nucleic acid preparations were extracted from leaves (60 mg)with phenol-chloroform, recovered by ethanol precipitation andresuspended in water (50 ll) (Rodio et al., 2007). When indicated,albino sectors from PC-expressing leaves were excised from thesurrounding green tissues with a razor blade. PLMVd accumulationin infected tissues was quantified by spotting aliquots (5 ll) of 1/5, 1/25 and 1/125 dilutions of the nucleic acid preparations ontopositively charged nylon membranes (Roche Diagnostics GmbH,http://www.roche.com) that were hybridized with a PLMVd-specificdigoxigenin-labeled riboprobe (Rodio et al., 2006).

Amplification and sequencing of PLMVd progeny variants

For cloning PLMVd-cDNAs from progeny variants, nucleic acidspreparations (150 ll) were obtained from leaf tissues (60 mg) by amodified silica-gel capture system (Foissac et al., 2001); whendealing with symptomatic leaves, albino and green sectors weredissected beforehand. Aliquots (5 ll) were used for synthesizingfirst-strand cDNA with random hexamers and the High CapacityReverse Transcription kit (Applied Biosystem, http://www.appliedbiosystems.com). The resulting cDNAs were PCR-amplified with theprimer pair FPLMV-57 (5¢-CACACCCCCCTCGGAACCAACCG-3¢) andFPLMV-58 (5¢-ATCCAGGTACCGCCGTAGAAAC-3¢), complementaryand identical to positions 202–180 and 203–224 of the referencePLMVd variant (Hernandez and Flores, 1992; Ambros et al., 1998),respectively, and the Expand High Fidelity PCR system (RocheDiagnostics GmbH). Amplicons of the expected size were cloned

in p-GEM-T Easy plasmid (Promega, http://www.promega.com)and sequenced (MWG-Biotech, http://www.mwg-biotech.com).The GenBank IDs for PLMVd progeny variants are: JN377825–JN377849; JN377851–JN377862; JN377864–JN377874; andJN377876–JN377891.

PLMVd infectious clones, site-directed mutagenesis and

inoculation of in vitro transcripts

Plasmids containing head-to-tail PLMVd-cDNA dimeric inserts fromnatural variants C40 (AJ550912) and P1.148 (DQ222050) have beendescribed previously (Malfitano et al., 2003; Rodio et al., 2006); fordescriptions of plasmids of variants C40(A349)-g12 (JN377855),P1.148(C349) (JN377863), C40(A349) (JN377850) and C40(stem)(JN377875), see Appendix S2. Recombinant plasmids were linearizedwith appropriate restriction enzymes and transcribed with T7 or SP6RNA polymerases. The resulting products were analyzed by elec-trophoresis in 5% polyacrylamide gels containing 1X TBE buffer and8 M urea, eluted and slash-inoculated into GF-305 peach seedlings.

Deep sequencing and bioinformatics analyses

The protocol for purifying the sRNAs, adaptor ligation, RT-PCRamplification, library purification and high-throughput DNAsequencing on the Illumina Genome Analyzer (Fasteris SA, http://www.fasteris.com), has been reported (Di Serio et al., 2010). Fourlibraries were sequenced. Two bar-coded leaf samples, from mock-inoculated and C40-infected GF-305 seedlings, were analyzed in asingle channel in the Illumina EAS269 GAII; the sRNA library fromthe C40-infected GF-305 seedling was generated from albino sectorsdissected from adjacent green tissues. The two additional bar-co-ded libraries from green leaf tissues of C40- and P1.148-infectedGF305 seedlings were sequenced in a single channel in the IlluminaGenome Analyzer HiSeq 2000. Raw data from the Illumina platformwere fed into an in-house pipeline, which removed barcodes andadaptors and split the clean sequences by size. Sequence sizesbetween 18 and 26 nt were blasted (BLASTN; Altschul et al., 1997)against a selected set of PLMVd variants, and a mapping profileagainst the consensus of the multiple alignment (CLUSTALW;Thompson et al., 1994) was generated. To find out potential targetsof PLMVd-sRNAs, sequences perfectly matching genomic RNAs ofC40 and its progeny were used for RNAhybrid (Rehmsmeier et al.,2004) searching on exon sequence fragments of the peach genome(Peach v1.0; International Peach Genome Initiative, http://www.rosaceae.org/peach/genome). The quality of the duplex pair-ing was estimated as proposed previously (Fahlgren and Carring-ton, 2010). Peach mRNAs detected as transcripts and formingpotential duplexes with PLMVd-sRNAs spanning totally or partially(in at least one nucleotide) the PC-associated insertion (PC-sRNAs)were further analysed by CHLOROP (http://www.cbs.dtu.dk/services/ChloroP/; Emanuelsson et al., 1999) to predict the presence ofchloroplastic transit peptides (Li and Chiu, 2010) in the encodedproteins.

Transcript analysis

Transcript levels were estimated by RT-PCR. Total RNAs (200 ng),obtained by treating total nucleic acid preparations with RQ1DNase I (Promega), were reverse transcribed with random hexa-mers. Aliquots (2 ll) of serial dilutions (1:2, 1:20 and 1:200) of theresulting cDNAs were added to amplification reactions (25 ll)catalyzed with Go-Taq DNA polymerase (Promega). The cyclingprogram, consisting in an initial denaturation at 94�C for 3 min and30 cycles (94�C for 30 s, 50�C for 30 s for 18S, psbA and RpoB, and55�C for cHSP90, and 72�C for 30 s), with a final extension at 72�C for7 min, was adopted according to preliminary experiments showing

1000 Beatriz Navarro et al.

ª 2012 The AuthorsThe Plant Journal ª 2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 991–1003

that cDNA amplification was in the logarithmic phase. A cDNAfragment (405 nt) of cHSP90 mRNA was RT-PCR amplified withprimers cHSP90-Fw (5¢-CAATGGCTCCAGTTCTAAGCA-3¢) andcHSP90-Rv (5¢-GCAGAGAGGGCTCAGTCACACTCAA-3¢), identicaland complementary, respectively, to positions 84–104 and 464–488of this transcript. Sequencing of the resulting product confirmed theexpected cDNA (JN377892). Primer pairs for cDNA amplification ofthe 18S rRNA and psbA and RpoB transcripts have been describedpreviously (Rodio et al., 2007).

5¢ RNA ligase-mediated rapid amplification of cDNA ends

(RLM-RACE)

RQ1 DNase I-treated total RNAs (1.2 lg) were incubated at 20�C for6 h with an RNA adaptor (5¢-GUUCAGAGUUCUACAGUCCGACG-AUC-3¢) and 0.5 U of T4 RNA ligase (Promega). Ligated RNAs werereverse transcribed with primer cHSP90-Rv as described before.First PCR was then performed using the forward primer P2 (5¢-AATGATACGGCGACCACCGACAGGTTCAGAGTTCTACAGTCCGA-3¢),with the 3¢ moiety (in bold) homologous to the RNA adaptor, and thereverse primer cHSP90-Rv reported above, whose 5¢ end maps288 nt downstream the cleavage site of peach cHSP90 mRNApredicted by PC-sRNA8a and PC-sRNA8b. The resulting product wasamplified with the same primer P2 and the nested reverse primercHSP90-nes-Rv (5¢-CCTTGTGGCTGTATAGACTATG-3¢), comple-mentary to positions 383–404 of peach cHSP90 mRNA. Followingelectrophoresis in a 1.2% agarose gel, the nested PCR product(248 bp) was excised, cloned into the pGEM-T easy vector (Pro-mega) and sequenced.

ACKNOWLEDGEMENTS

This work was supported by a dedicated grant from the ItalianMinistry of Economy and Finance to the CNR (Legge n. 191/2009),the Dipartimento Agroalimentare of the CNR of Italy (A. Leone andD. Mariotti 2008 award for advanced research in agriculture to FDS)and by the Ministerio de Ciencia e Innovacion of Spain (grantsBFU2008-03154 and BFU2011-28443 to RF).

SUPPORTING INFORMATION

Additional Supporting Information may be found in the onlineversion of this article:Figure S1. Peach phenotype and progeny of the natural PLMVdvariant C40(A349)-g12.Figure S2. Relative distribution by size classes of total sRNAs fromGF-305 peach leaves infected by variant C40 (blue) or mockinoculated (red).Figure S3. Distribution by size classes, polarity and abundance ofPLMVd-sRNAs in the C40-infected sample.Figure S4. Location and frequency of the 5¢ termini of PLMVd-sRNAsalong the genomic RNAs.Figure S5. Number and location of the 5¢ termini of 21-nt non-redundant (nr) PLMVd-sRNAs along the genomic RNAs (a), andnucleotide variability detected in variant C40 and its progeny (b).Figure S6. Distribution by size classes, polarity and abundance ofPLMVd-sRNAs in the green tissue of C40- and P1.148-infectedsamples.Figure S7. Duplexes potentially formed by the cHSP90 mRNA andPLMVd-sRNAs with hairpin insertions obtained by deep sequencingof sRNA libraries from green leaves of C40- and P1.148-infectedpeach seedlings (a and b, respectively).Table S1. Predicted peach mRNAs targeted by PLMVd-sRNAsmatching perfectly the genomic sequence of variant C40 and itsprogeny.

Table S2. Chloroplast transit peptides and their length in cHSP90 offive species predicted by the CHLOROP program.Appendix S1. PLMVd-sRNAs (18–26 nt) from albino leaf sectorsinfected with PLMVd variant C40.Appendix S2. Experimental procedures.Please note: As a service to our authors and readers, this journalprovides supporting information supplied by the authors. Suchmaterials are peer-reviewed and may be re-organized for onlinedelivery, but are not copy-edited or typeset. Technical supportissues arising from supporting information (other than missingfiles) should be addressed to the authors.

REFERENCES

Altschul, S.F., Madden, T.L., Schaffer, A.A., Zhang, J., Zhang, Z., Miller, W.

and Lipman, D.J. (1997) Gapped BLAST and PSI-BLAST: a new generation

of protein database search programs. Nucleic Acids Res. 25, 3389–3402.

Ambros, S., Hernandez, C., Desvignes, J.C. and Flores, R. (1998) Genomic

structure of three phenotypically different isolates of peach latent mosaic

viroid: implications of the existence of constraints limiting the heteroge-

neity of viroid quasi-species. J. Virol. 72, 7397–7406.

Ambros, S., Hernandez, C. and Flores, R. (1999) Rapid generation of genetic

heterogeneity in progenies from individual cDNA clones of peach latent

mosaic viroid in its natural host. J. Gen. Virol. 80, 2239–2252.

Bolduc, F., Hoareau, C., St-Pierre, P. and Perreault, J.P. (2010) In-depth

sequencing of the siRNAs associated with peach latent mosaic viroid

infection. BMC Mol. Biol. 11, 16.

Bussiere, F., Lehoux, J., Thompson, D.A., Skrzeczkowski, L.J. and Perreault,

J. (1999) Subcellular localization and rolling circle replication of

peach latent mosaic viroid: hallmarks of group A viroids. J. Virol. 73,

6353–6360.

Cao, D., Froehlich, J.E., Zhang, H. and Cheng, C.L. (2003) The chlorate-resis-

tant and photomorphogenesis-defective mutant cr88 encodes a chloro-

plast-targeted HSP90. Plant J. 33, 107–118.

Carbonell, A., Martınez de Alba, A.E., Flores, R. and Gago, S. (2008)

Double-stranded RNA interferes in a sequence-specific manner with

infection of representative members of the two viroid families. Virology,

371, 44–53.

Carthew, R.W. and Sontheimer, E.J. (2009) Origins and Mechanisms of

miRNAs and siRNAs. Cell, 136, 642–655.

Chen, X. (2009) Small RNAs and their roles in plant development. Annu. Rev.

Cell Dev. Biol. 25, 21–44.

Chen, B., Zhong, D. and Monteiro, A. (2006) Comparative genomics and

evolution of the HSP90 family of genes across all kingdoms of organisms.

BMC Genomics, 7, 156.

Csorba, T., Pantaleo, V. and Burgyan, J. (2009) RNA silencing: an antiviral

mechanism. Adv. Virus Res. 75, 35–71.

De la Pena, M., Navarro, B. and Flores, R. (1999) Mapping the molecular

determinant of pathogenicity in a hammerhead viroid: a tetraloop within

the in vivo branched RNA conformation. Proc. Natl Acad. Sci. USA, 96,

9960–9965.

Di Serio, F., Gisel, A., Navarro, B., Delgado, S., Martınez de Alba, A.E.,

Donvito, G. and Flores, R. (2009) Deep sequencing of the small RNAs de-

rived from two symptomatic variants of a chloroplastic viroid: implications

for their genesis and for pathogenesis. PLoS ONE, 4, e7539.

Di Serio, F., Martınez de Alba, A.E., Navarro, B., Gisel, A. and Flores, R.

(2010) RNA-dependent RNA polymerase 6 delays accumulation and

precludes meristem invasion of a nuclear-replicating viroid. J. Virol. 84,

2477–2489.

Diener, T.O. (1972) Potato spindle tuber viroid. VIII. Correlation of infectivity

with a UV-absorbing component and thermal denaturation properties of

the RNA. Virology, 50, 606–609.

Diermann, N., Matousek, J., Junge, M., Riesner, D. and Steger, G. (2010)

Characterization of plant miRNAs and small RNAs derived from potato

spindle tuber viroid (PSTVd) in infected tomato. Biol. Chem. 391, 1379–

1390.

Ding, B. (2009) The biology of viroid-host interactions. Annu. Rev. Phytopa-

thol. 47, 105–131.

Ding, S.W. (2010) RNA-based antiviral immunity. Nat. Rev. Immunol. 10, 632–

644.

Viroid-induced RNA silencing of host mRNA 1001

ª 2012 The AuthorsThe Plant Journal ª 2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 991–1003

Dunoyer, P. and Voinnet, O. (2009) Movement of RNA silencing between

plant cells: is the question now behind us? Trends Plant Sci. 14, 643–

644.

Emanuelsson, O., Nielsen, H. and von Heijne, G. (1999) ChloroP, a neural

network-based method for predicting chloroplast transit peptides and their

cleavage sites. Protein Sci. 8, 978–984.

Fahlgren, N. and Carrington, J.C. (2010) miRNA target prediction in plants.

Methods Mol. Biol. 592, 51–57.

Flores, R., Daros., J.A. and Hernandez, C. (2000) The Avsunviroidae family:

viroids with hammerhead ribozymes. Adv. Virus Res. 55, 271–323.

Flores, R., Hernandez, C., Martınez de Alba, A.E., Daros., J.A. and Di Serio, F.

(2005) Viroids and viroid-host interactions. Annu. Rev. Phytopathol. 43,

117–139.

Flores, R., Delgado, S., Rodio, M.E., Ambros, S., Hernandez, C. and Di Serio, F.

(2006) Peach latent mosaic viroid: not so latent. Mol. Plant Pathol. 7, 209–

221.

Foissac, X., Svanella-Dumas, L., Dulucq, M.J., Gentit, P. and Candresse, T.

(2001) Polyvalent detection of fruit tree tricho, capillo and foveaviruses by

nested RT-PCR using degenerated and inosine containing primers (PDO

RT-PCR). Acta Hortic. 550, 37–43.

Gago, S., De la Pena, M. and Flores, R. (2005) A kissing-loop interaction in a

hammerhead viroid RNA critical for its in vitro folding and in vivo viability.

RNA, 11, 1073–1083.

Gago, S., Elena, S.F., Flores, R. and Sanjuan, R. (2009) Extremely high muta-

tion rate of a hammerhead viroid. Science, 323, 1308.

Gomez, G., Martınez, G. and Pallas, V. (2008) Viroid-induced symptoms in

Nicotiana benthamiana plants are dependent on RDR6 activity. Plant

Physiol. 148, 414–423.

Gross, H.J., Domdey, H., Lossow, C., Jank, P., Raba, M., Alberty, H. and

Sanger, H.L. (1978) Nucleotide sequence and secondary structure of potato

spindle tuber viroid. Nature, 273, 203–208.

Hernandez, C. and Flores, R. (1992) Plus and minus RNAs of peach latent

mosaic viroid self-cleave in vitro through hammerhead structures. Proc.

Natl Acad. Sci. USA, 89, 3711–3715.

Hou, W.N., Duan, C.G., Fang, R.X., Zhou, X.Y. and Guo, H.S. (2011) Satellite

RNA reduces expression of the 2b suppressor protein resulting in the

attenuation of symptoms caused by cucumber mosaic virus infection. Mol.

Plant Pathol. 12, 595–605.

Hutchins, C., Rathjen, P.D., Forster, A.C. and Symons, R.H. (1986) Self-cleav-

age of plus and minus RNA transcripts of avocado sunblotch viroid. Nucleic

Acids Res. 14, 3627–3640.

Itaya, A., Folimonov, A., Matsuda, Y., Nelson, R.S. and Ding, B. (2001) Potato

spindle tuber viroid as inducer of RNA silencing in infected tomato. Mol.

Plant-Microbe Interact. 14, 1332–1334.

Itaya, A., Matsuda, Y., Gonzales, R.A., Nelson, R.S. and Ding, B. (2002) Potato

spindle tuber viroid strains of different pathogenicity induces and sup-

presses expression of common and unique genes in infected tomato. Mol.

Plant-Microbe Interact. 15, 990–999.

Itaya, A., Zhong, X., Bundschuh, R., Qi, Y., Wang, Y., Takeda, R., Harris, A.R.,

Molina, C., Nelson, R.S. and Ding, B. (2007) A structured viroid RNA is

substrate for Dicer-like cleavage to produce biologically active small RNAs

but is resistant to RISC-mediated degradation. J. Virol. 81, 2980–2994.

Jay, F., Wang, Y., Yu, A., Taconnat, L., Pelletier, S., Colot, V., Renou, J.P. and

Voinnet, O. (2011) Misregulation of AUXIN RESPONSE FACTOR 8 underlies

the developmental abnormalities caused by three distinct viral silencing

suppressors in Arabidopsis. PLoS Pathog. 7, e1002035.

Kasschau, K.D., Xie, Z., Allen, E., Llave, C., Chapman, E.J., Krizan, K.A. and

Carrington, J.C. (2003) P1/HC-Pro, a viral suppressor of RNA silencing,

interferes with Arabidopsis development and miRNA unction. Dev. Cell, 4,

205–217.

Li, H.M. and Chiu, C.C. (2010) Protein transport into chloroplasts. Annu. Rev.

Plant Biol. 61, 157–180.

Llave, C. (2010) Virus-derived small interfering RNAs at the core of plant-virus

interactions. Trends Plant Sci. 15, 701–707.

Malfitano, M., Di Serio, F., Covelli, L., Ragozzino, A., Hernandez, C. and Flores,

R. (2003) Peach latent mosaic viroid variants inducing peach calico contain

a characteristic insertion that is responsible for this symptomatology.

Virology, 313, 492–501.

Mallory, A. and Vaucheret, H. (2010) Form, function, and regulation of ARG-

ONAUTE proteins. Plant Cell, 22, 3879–3889.

Mallory, A.C., Reinhart, B.J., Jones-Rhoades, M.W., Tang, G., Zamore, P.D.,

Barton, M.K. and Bartel, D.P. (2004) MicroRNA control of PHABULOSA in

leaf development: importance of pairing to the microRNA 5¢ region. EMBO

J. 23, 3356–3364.

Markarian, N., Li, H.W., Ding, S.W. and Semancik, J.S. (2004) RNA

silencing as related to viroid induced symptom expression. Arch. Virol.

149, 397–406.

Martınez de Alba, A.E., Flores, R. and Hernandez, C. (2002) Two chloroplastic

viroids induce the accumulation of the small RNAs associated with post-

transcriptional gene silencing. J. Virol. 76, 13094–13096.

Martınez, G., Donaire, L., Llave, C., Pallas, V. and Gomez, G. (2010) High-

throughput sequencing of Hop stunt viroid-derived small RNAs from

cucumber leaves and phloem. Mol. Plant Pathol. 11, 347–359.

Matousek, J., Kozlova, P., Orctova, L., Schmitz, A., Pesina, K., Bannach, O.,

Diermann, N., Steger, G. and Riesner, D. (2007) Accumulation of viroid-

specific small RNAs and increase of nucleolytic activities linked to viroid-

caused pathogenesis. Biol. Chem. 388, 1–13.

Meinke, D., Muralla, R., Sweeney, C. and Dickerman, A. (2008) Identifying

essential genes in Arabidopsis thaliana. Trends Plant Sci. 13, 483–491.

Meyer, P., Prodromou, C., Hu, B., Vaughan, C., Roe, S.M., Panaretou, B., Piper,

P.W. and Pearl, L.H. (2003) Structural and functional analysis of the middle

segment of hsp90: implications for ATP hydrolysis and client protein and

cochaperone interactions. Mol. Cell, 11, 647–658.

Moissiard, G. and Voinnet, O. (2006) RNA silencing of host transcripts by cau-

liflower mosaic virus requires coordinated action of the four Arabidopsis

Dicer-like proteins. Proc. Natl Acad. Sci. USA, 103, 19593–19598.

Navarro, B. and Flores, R. (1997) Chrysanthemum chlorotic mottle

viroid: unusual structural properties of a subgroup of self-cleaving viroids

with hammerhead ribozymes. Proc. Natl Acad. Sci. USA, 94, 11262–11267.

Navarro, B., Pantaleo, V., Gisel, A., Moxon, S., Dalmay, T., Bisztray, G.,

Di Serio, F. and Burgyan, J. (2009) Deep sequencing of viroid-derived small

RNAs from grapevine provides new insights on the role of RNA silencing in

plant-viroid interaction. PLoS ONE, 4, e7686.

Papaefthimiou, I., Hamilton, A.J., Denti, M.A., Baulcombe, D.C., Tsagris, M.

and Tabler, M. (2001) Replicating potato spindle tuber viroid RNA is

accompanied by short RNA fragments that are characteristic of post-tran-

scriptional gene silencing. Nucleic Acids Res. 29, 2395–2400.

Rehmsmeier, M., Steffen, P., Hochsmann, M. and Giegerich, R. (2004) Fast

and effective prediction of microRNA/target duplexes. RNA, 10, 1507–1517.

Rodio, M.E., Delgado, S., Flores, R. and Di Serio, F. (2006) Variants of peach

latent mosaic viroid inducing peach calico: uneven distribution in infected

plants and requirements of the insertion containing the pathogenicity

determinant. J. Gen. Virol. 87, 231–240.

Rodio, M.E., Delgado, S., De Stradis, A.E., Gomez, M.D., Flores, R. and

Di Serio, F. (2007) A viroid RNA with a specific structural motif inhibits

chloroplast development. Plant Cell, 19, 3610–3626.

Schmitz, G., Schmidt, M. and Feierabend, J. (1996) Characterization of a

plastid-specific HSP90 homologue: identification of a cDNA sequence,

phylogenetic descendence and analysis of its mRNA and protein expres-

sion. Plant Mol. Biol. 30, 479–492.

Schroda, M. and Muhlhaus, T. (2009) A ‘foldosome’ in the chloroplast? Plant

Signal Behav. 4, 301–303.

Schwind, N., Zwiebel, M., Itaya, A., Ding, B., Wang, M.B., Krczal, G. and

Wassenegger, M. (2009) RNAi-mediated resistance to potato spindle tuber

viroid in transgenic tomato expressing a viroid hairpin RNA construct. Mol.

Plant Pathol. 10, 459–469.

Shimura, H., Pantaleo, V., Ishihara, T., Myojo, N., Inaba, J., Sueda, K., Burg-

yan, J. and Masuta, C. (2011) A viral satellite RNA induces yellow symp-

toms on tobacco by targeting a gene involved in chlorophyll biosynthesis

using the RNA silencing machinery. PLoS Pathog. 7, e1002021.

Smith, N.A., Eamens, A.L. and Wang, M.B. (2011) Viral small interfering RNAs

target host genes to mediate disease symptoms in plants. PLoS Pathog. 7,

e1002022.

St-Pierre, P., Hassen, F., Thompson, D. and Perreault, J.P. (2009) Character-

ization of the siRNAs associated with peach latent mosaic viroid infection.

Virology, 383, 178–182.

Symons, R.H. (1981) Avocado sunblotch viroid: primary sequence and pro-

posed secondary structure. Nucleic Acids Res. 9, 6527–6537.

Tessitori, M., Maria, G., Capasso, C., Catara, G., Rizza, S., De Luca, V., Catara,

A., Capasso, A. and Carginale, V. (2007) Differential display analysis of gene

1002 Beatriz Navarro et al.

ª 2012 The AuthorsThe Plant Journal ª 2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 991–1003

expression in Etrog citron leaves infected by citrus viroid III. Biochim.

Biophys. Acta, 1769, 228–235.

Thompson, J.D., Higgins, D.G. and Gibson, T.J. (1994) CLUSTAL W: improv-

ing the sensitivity of progressive multiple sequence alignment through

sequence weighting, position-specific gap penalties and weight matrix

choice. Nucleic Acids Res. 22, 4673–4680.

Tsagris, E.M., Martınez de Alba, A.E., Gozmanova, M. and Kalantidis, K. (2008)

Viroids. Cell. Microbiol. 10, 2168–2179.

Vogt, U., Pelissier, T., Putz, A., Razvi, F., Fischer, R. and Wassenegger, M.

(2004) Viroid-induced RNA silencing of GFP-viroid fusion transgenes does

not induce extensive spreading of methylation or transitive silencing. Plant

J. 1, 107–118.

Voinnet, O. (2008) Use, tolerance and avoidance of amplified RNA silencing

by plants. Trends Plant Sci. 13, 317–328.

Wang, M.B., Bian, X.Y., Wu, L.M. et al. (2004) On the role of RNA silencing in

the pathogenicity and evolution of viroids and viral satellites. Proc. Natl

Acad. Sci. USA, 101, 3275–3280.

Wang, Y., Shibuya, M., Taneda, A., Kurauchi, T., Senda, M., Owens, R.A. and

Sano, T. (2011) Accumulation of potato spindle tuber viroid-specific small

RNAs is accompanied by specific changes in gene expression in two

tomato cultivars. Virology, 413, 72–83.

Willmund, F. and Schroda, M. (2005) HEAT SHOCK PROTEIN 90C is a bona

fide Hsp90 that interacts with plastidic HSP70B in Chlamydomonas rein-

hardtii. Plant Physiol. 138, 2310–2322.

Willmund, F., Dorn, K.V., Schulz-Raffelt, M. and Schroda, M. (2008) The

chloroplast DnaJ homolog CDJ1 of Chlamydomonas reinhardtii is part of a

multichaperone complex containing HSP70B, CGE1, and HSP90C. Plant

Physiol. 148, 2070–2082.

Viroid-induced RNA silencing of host mRNA 1003

ª 2012 The AuthorsThe Plant Journal ª 2012 Blackwell Publishing Ltd, The Plant Journal, (2012), 70, 991–1003