C.R. Acad. Sci.

Paris, Sciences de la vie / Life Sciences 324 (2001) 943952

2001 Acadmie des sciences/ditions scientifiques et mdicales. Tous droits rservs
S0764446901013701/FLA

A naked plant-specific RNA ten-fold smaller than

the smallest known viral RNA: the viroid
Ricardo Flores*
Instituto de Bologa Molecular y Celular de Plantas (UPV-CSIC), Universidad Politcnica de Valencia, Avenida
de los Naranjos s/n, 46022, Valencia, Spain
Received 30 April 2001; accepted 7 May 2001
Communicated by Josy Bov

Abstract Viroids are subviral plant pathogens at the frontier of life. They are solely
composed by a single-stranded circular RNA of 246401 nt with a compact secondary
structure. Viroids replicate autonomously when inoculated into their host plants and
incite, in most of them, economically important diseases. In contrast to viruses, viroids
do not code for any protein and depend on host enzymes for their replication, which in
some viroids occurs in the nucleus and in others in the chloroplast, through a rolling-
circle mechanism with three catalytic steps. Quite remarkably, however, one of the steps,
cleavage of the oligomeric head-to-tail replicative intermediates to unit-length strands, is
mediated in certain viroids by hammerhead ribozymes that can be formed by their
strands of both polarities. Viroids induce disease by direct interaction with host factors,
the nature of which is presently unknown. Some properties of viroids, particularly the
presence of ribozymes, suggest that they might have appeared very early in evolution
and could represent living fossils of the precellular RNA world that presumably
preceded our current world based on DNA and proteins. 2001 Acadmie des
sciences/ditions scientifiques et mdicales

hammerhead ribozyme / viroid

Rsum Virode : un ARN nu, spcifique de plante, dix fois plus petit que
le plus petit ARN viral connu. Les virodes sont des pathognes subviraux des plan-
tes la frontire de la vie. Ils sont seulement composs dun ARN circulaire monocat-
naire de 246401 nt ayant une structure secondaire compacte. Les virodes se repliquent
de faon autonome quand ils sont inoculs leurs plantes htes et induisent dans la
plupart de cettes plantes des maladies dimportance conomique. Au contraire des virus,
les virodes ne codent pour aucune proteine et dpendent des enzymes de leur htes
pour leur rplication., qui pour certains virodes a lieu dans le noyau et pour dautres
dans le chloroplaste, suivant un mchanisme de cercle roulant avec trois tapes
catalitiques. Cependant, de faon remarquable, une des ces tapes, la coupure des
intermdiares rplicatifs oligomriques ttequeue en units monomriques, est mdie
chez certains virodes par des ribozymes de tte de marteau qui peuvent tre forms par
leurs chanes de chaque polarit. Les virodes induisent les maladies par interactions
directes avec des facteurs de lhte dont la nature nest pas connue. Quelques propriets
des virodes, en particulier la prsence des ribozymes, suggrent quils pourraient avoir
apparu trs tt lors de lvolution et pourraient reprsenter des fossiles vivants du
monde prcellulaire ARN qui a probablement prcd notre monde actuel bas sur
lADN et les protines. 2001 Acadmie des sciences/ditions scientifiques et mdicales
ribozyme tte de marteau / virode

*Correspondence and reprints.

E-mail address: rflores@ibmcp.upv.es (R. Flores).

943
 R. Flores / C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 324 (2001) 943952
. version abrge
Le concept de virode a merg et sest affirm il y a
environ trente ans la suite dtudes destines car-
actriser les agents causaux de certaines maladies des
plantes dont on pensait initialement quelles taient
dues des virus. Le virode des tubercules fusiformes
de la pomme de terre (Potato spindle tuber viroid,
PSTVd) a t le premier tre identifi et caractris
parmi les vingt-huit virodes connus lheure actuelle.
Les virodes sont constitus dune chane dARN mono-
catnaire de 246 401 rsidus nuclotidiques qui, dans
la plupart des cas, adopte une structure secondaire
compacte. Chez les vingt-cinq virodes de la famille des
Pospiviroidae (espce type : PSTVd), on a observ
dans cette structure une srie de motifs conservs dont
le plus prominent est une rgion centrale conserve
(CCR, pour central conserved region). Ces motifs
nexistent pas chez les trois autres virodes, ceux de la
famille des Avsunviroidae (espce type : virode des
taches solaires de lavocat (Avocado sunblotch viroid,
ASBVd), mais dont la chane plus comme la chane
moins ont la proprit remarquable de sautocliver
par activit du ribozyme tte de marteau (cette
activit nexiste pas chez les virodes de la premire
famille). Un autre caractre de diffrenciation rside
dans le fait que le PSTVd se rplique et saccumule
dans le noyau, alors que ces phnomnes se droulent
dans le chloroplaste pour le virode ASBVd (les autres
virodes se comportent comme lespce type de leur
famille).
Les virodes, comme les virus, se rpliquent dune
faon autonome quand on les inocule une plante-
hte, mais contrairement aux virus, ils ne codent pour
aucune protine. La rplication des virodes se fait par
un mcanisme de cercle roulant en trois tapes
catalyses respectivement par : 1) une ARN polymrase
qui, par transcription ritrative ARNARN, produit des
chanes oligomriques, 2) une RNase qui coupe les
oligomres en chanes monomriques et 3) une ARN
ligase pour circulariser les monomres. Soit lARN
oligomrique moins qui rsulte de la premire tran-
scription ARNARN, est utilis directement comme
matrice pour la seconde transcription ARNARN, soit il
est coup puis ligatur pour donner des monomres
circulaires moins , alors la rplication est asymtrique
ou symtrique. Le PSTVd et les autres membres de sa
1. Introduction: a historical perspective
The foundations of microbiology as an independent
scientific discipline, were laid during the second half of
the XIX century. The pioneering work of Pasteur and Koch
established the bacterial nature of a series of infectious
944
famille suivent la voie asymtrique, alors que le ASBVd
et les deux autres membres de la famille empruntent la
voie symtrique. Chez les Pospiviroidae la premire
tape est catalyse par lARN polymrase II du noyau,
alors que chez lASBVd, lenzyme implique semble
tre un ARN polymrase chloroplastique encode par
le noyau. Chez lASBVd et peut-tre chez tous les
virodes, linitiation de la synthse de lARN commence
des sites spcifiques, suggrant fortement que la
transcription serait promoteur-dpendant. Lactivit
RNase de la deuxime tape est celle dune protine
code par la plante-hte chez les Pospiviroidae, mais
chez les Avsunviroidae, il sagit d une structure de
ribozyme en forme de tte de marteau aussi bien dans
la chane moins que la chane plus . LARN ligase
implique dans la troisime tape est vraisemblable-
ment une enzyme de la plante-hte, bien quelle ait t
propos que cette tape puisse tre autocatalytique
chez au moins un virode.
Les virodes causent une srie de maladies chez des
plantes conomiquement importantes comme la
pomme de terre, la tomate, le concombre, le pcher, le
pommier, le poirier, le prunier, les agrumes, et des
plantes ornementales comme le chrysanthme et le
colus. Certains virodes ont des effets dvastateurs,
comme celui de la maladie du cadang-cadang du
cocotier, qui est responsable de la mort de plus de
20 millions darbres aux Philippines. Dautres virodes
ninduisent pas de symptmes ou seulement des lgers.
La propagation des virodes dans la nature est le
rsultat de pratiques agricoles (utilisation de matriel
vgtal contamin, doutils de taille contamins).
lintrieur de la plante, les virodes semblent utiliser les
plasmodesmes pour leurs mouvements de cellule
cellule et le phlome pour les mouvements plus
longue distance. Puisque les virodes ont des gnomes
non-codant, ils doivent induire leurs effets pathognes
par interaction directe avec des facteurs de la plante-
hte, dont la nature nest pas connue aujourdhui.
Les virodes semblent avoir une origine volutive
indpendante de celle des virus. Certaines proprits
des virodes, en particulier la prsence de ribozymes,
suggrent quils sont apparus trs tt au cours de
lvolution et quils pourraient reprsenter des fossiles
vivants du monde prcellulaire ARN qui a vraisem-
blablement prcd notre monde actuel bas sur lADN
et les protines.
diseases. However, by the end of his life, Pasteur had not
succeeded in his attempts to apply the methodology that
had allowed him to grow and identify several microbial
pathogens to the case of the agent of rabies. The reason for
this failure was that he was dealing with a pathogen
radically different from those so far known. A proposal for
 R. Flores / C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 324 (2001) 943952
the existence of a new category of much simpler patho-
gens came at the turn of the XIX and XX century as a
consequence of the experiments by Mayer, Ivanosky and
Beijerinck, aimed at identifying the agent of tobacco
mosaic disease [1]. More specifically, Beijerinck noted
that the agent, which he called virus, multiplied only in the
living plant and not in in vitro media, implicitly advancing
the existence of a new type of hostpathogen interaction
(viruses are now known to be strictly intracellular parasites
whose replication depends on the metabolism of the host
cell that they infect). Later, viruses were found to infect not
only plants but all types of cells, including those of animal,
fungal and bacterial origin, and even simple mycoplasm
cells. Viruses were thus considered the smallest biological
entity, a paradigm remaining undisputed for decades. It is
not surprising that the concept of viroid, a term coined by
Diener to name the agent of potato spindle tuber (PST)
disease and the first example of a singular class of autono-
mous replicating subviral pathogens [2], was received
with some skepticism by the scientific community. But
experimental evidence gathered over the next ten years
eroded any skepticism. As an example of this and follow-
ing a debate with Diener in the International Congress of
Virology (Strasbourg, 1981), Andr Lwoff, one of the most
respected authorities in virology, stated that he was con-
vinced of the existence of viroids. Victor Hugo wrote:
Une invasion darmes peut tre resiste mais non pas
une ide lorsque son temps est arriv (An invasion of
armies can be resisted, but not an idea whose time has
come). The time had come for viroids. They are currently
regarded as the lowest step of the biological scale: molecu-
lar parasites at the frontier of life. Interestingly enough, like
tobacco mosaic virus (TMV) was the first virus isolated and
characterized, potato spindle tuber viroid (PSTVd) has
been the first independently replicating subviral pathogen
identified. Both examples illustrate that the study of plant
systems has enriched the history of biology with a plethora
of new and exciting discoveries.
2. Viroids: unique structural
and functional properties
The viroid concept emerged and was reinforced as a
result of researches aimed at characterizing the agent of
PST disease. This disease was initially thought to be
induced by a virus but experiments aimed at concentrating
and/or separating the presumed viral particles produced
some unexpected results: after differential centrifugation
most of the infectious principle remained in the superna-
tant and after rate zonal centrifugation it migrated into
regions of the gradient corresponding to entities consider-
ably smaller than conventional virions. Moreover, similar
observations were made when the extracts were pre-
treated with a chemical for removing proteins such as
phenol, indicating that the infectious principle was a
naked nucleic acid. Additional experiments using poly-
acrylamide gel electrophoresis (PAGE), a separative tech-
nique which resolves complex mixtures of nucleic acids,
showed that the infectious principle was an unusually
small nucleic acid. This small nucleic acid turned out to be
an RNA, because it was resistant to DNase but sensitive to
RNase, and presumably had a circular structure because
treatments with exonucleases did not affect its infectivity. It
must be noted here that all these studies used tomato
seedlings as an experimental host; this choice eased the
experimental work considerably as the tomato is effort-
lessly grown in the greenhouse and reacts to infection with
diagnostic symptoms within a relative short time
(1214 days). There was a chance that the small circular
RNA associated with the PST disease could be a satellite
RNA functionally dependent on a helper virus, but
repeated attempts to detect viral particles in infected tis-
sues failed. Furthermore, the hypothetical helper virus
should have an extremely efficient seed transmission in
tomato, and it should also infect other alternative experi-
mental hosts known. Finally, the possibility that the infec-
tious PST-associated RNA could actually be a population
of similarly sized RNAs with different sequences, collec-
tively representing the equivalent to a conventional virus
genome, was dismissed when RNase fingerprint analysis
revealed a structural complexity consistent with a single
molecule. Therefore, it had to be concluded that the
infectious principle of PST disease was indeed a small
circular RNA endowed with autonomous replication (a
viroid) [3].
Definitive support for the new concept was obtained
when the viroid RNA was identified as a physical entity:
the infectivity of RNA preparations separated by PAGE
was associated with a differential UV-absorbing peak
absent in healthy controls (figure 1) [4]. Also, electron
microscopy of purified PSTVd enabled direct visualization
and measurement of its small size and provided evidence
supporting a rod-like secondary structure (figure 2) [5, 6].
PSTVd was just the first player in this new saga, and soon
after were discovered citrus exocortis viroid (CEVd) [7]
and then many others (table I) [8]. Table I shows that viroid
RNAs range between 246 and 401 nucleotide residues
(nt), which is approximately ten-fold smaller than the
smallest known viral RNA.
Differences between viruses and viroids are not con-
fined to genome size. Viroids, in contrast to viruses: 1) lack
the ability to code for proteins, a property with deep
implications on viroid replication and pathogenesis, 2) in
some cases have catalytic activity, i.e. express ribozyme
activity, and 3) may be relics of the precellular RNA world
that presumably preceded the world based on DNA and
proteins as we know it (see below).
PSTVd was the first viroid sequenced [9]. This piece of
work, a milestone in virology, confirmed the circular struc-
ture of the RNA and its high self-complementarity that
permits the adoption in vitro, and presumably in vivo, of a
rod-like secondary structure in which double-stranded
regions are separated by loops formed by unpaired resi-
dues. With the accelerated progress of molecular tech-
nologies, cloning and sequencing of new viroids has
945
 R. Flores / C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 324 (2001) 943952

Figure 2. Electron micrograph of a mixture of purified preparations of

Figure 1. UV-absortion profiles (continuous lines) of RNA prepara-
PSTVd RNA and viral T7 DNA. The size differences and the rod-like
tions from healthy (A) and PSTVd-infected (B) tomato leaves after
structure of the viroid are clearly illustrated. (Reproduced with per-
electrophoresis in 20 % polyacrylamide gels. Also presented in (B),
mission from Diener, 1979).
the infectivity distribution (broken line) with the infectivity index
(right) expressed in arbitrary units. 5S corresponds to 5S ribosomal
RNA, I, III and IV, to unidentified minor components of cellular RNA,
and II to PSTVd. A260 refers to absorbance at 260 nm. Electrophoretic
migration is from right to left. (Reproduced with permission from
Diener, 1971). sunblotch viroid (ASBVd) [12, 13], peach latent mosaic
viroid (PLMVd) [14] and chysanthemum chlorotic mottle
viroid (CChMVd) [15], have been allocated to a second
family, Avsunviroidae, on the basis of the absence of these
conserved motifs and also on the striking property that
become much easier and the group has presently expanded
their strands of both polarities are able to self-cleave
to a total number of 28 (table I). Comparative sequence
through hammerhead ribozymes (absent in members of
and structural analysis has shown that in addition to
the other family) (figure 3). Moreover, PLMVd and CCh-
PSTVd, most viroid species can also adopt a rod- or quasi
MVd adopt branched instead of rod-like conformations in
rod-like secondary structure with five domains (figure 3)
vitro and most likely in vivo. Differences between both
[10, 11]. Within this structure, three characteristic motifs
viroid families also include the subcellular site of replica-
have been identified: 1) the central conserved region
tion and accumulation (see below).
(CCR) formed by two sets of residues in the upper and
lower strands, 2) the terminal conserved region (TCR)
located in the upper strand of the left terminal domain, and 3. Viroids are important
3) the terminal conserved hairpin (TCH) also found at the
end of the left terminal domain and conserved not only in phytopathological agents
sequence but also in secondary structure (figure 3). The
sequence of the CCR and the presence or absence of the That viroids are agriculturally relevant is reflected by the
TRC and TCH motifs have served to group 25 viroid fact that PSTVd, and then CEVd, were discovered as a
species in a first family, Pospiviroidae, and, within it, in consequence of research work designed to identify the
different genera (table I). However, three viroids, avocado agents of two plant diseases. Since then, additional viroids

946
 R. Flores / C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 324 (2001) 943952

Table I. Viroids sequenced with their English names, abbreviations, size of typical sequence variants, and genus and family to which they belong.
Family Genus Viroid species Abbreviation Size (nt)
Pospiviroidae Pospiviroid potato spindle tuber PSTVd 356, 359360
tomato chlorotic dwarf TCDVd 360
Mexican papita MPVd 359360
tomato planta macho TPMVd 360
citrus exocortis CEVd 370375, 463
chrysanthemum stunt CSVd 354, 356
tomato apical stunt TASVd 360, 363
iresine 1 IrVd-1 370
columnea latent CLVd 370, 372
Hostuviroid hop stunt HSVd 295303
Cocadviroid coconut cadang-cadang CCCVd 246247, 287301
coconut tinangaja CTiVd 254
hop latent HLVd 256
citrus IV CVd-IV 284
Apscaviroid apple scar skin ASSVd 329330
citrus III CVd-III 294, 297
apple dimple fruit ADFVd 306307
grapevine yellow speckle 1 GVYSd-1 366368
grapevine yellow speckle 2 GVYSd-2 363
citrus bent leaf CBLVd 318
pear blister canker PBCVd 315316
Australian grapevine AGVd 369
Coleviroid coleus blumei 1 CbVd-1 248, 250251
coleus blumei 2 CbVd-2 301302
coleus blumei 3 CbVd-3 361362, 364
Avsunviroidae Avsunviroid avocado sunblotch ASBVd 246250
Pelamoviroid peach latent mosaic PLMVd 335338
chrysanthemum chlorotic mottle CChMVd 398401

have been characterized as the etiologic agents of a series
of maladies affecting economically important herbaceous
and woody plants such as potato, tomato, cucumber, hop,
citrus, coconut, grapevine, several subtropical and tem-
perate fruit trees including avocado, peach, apple, pear
and plum, and ornamentals such as chrysanthemum and
coleus (table I). Some viroids have devastating effects, as
illustrated by the case of coconut cadang-cadang viroid
(CCCVd) which has killed more than 20 million trees in
Southeast Asia (Philippines), while others cause epinasty,
rugosity, chlorosis and necrosis on leaves, internode short-
ening of stems leading to stunted plants, bark cracking,
deformations and color alterations of fruits and reserve
organs, and delays in foliation, flowering and ripening. A
few viroids induce very mild symptoms or no symptoms at
all. Viroid replication and symptom expression are gener-
ally favored by high temperatures and light intensities; it is
therefore not surprising that viroids affect mainly crops
grown in tropical or subtropical areas and in greenhouses.
The host range of different viroids is very variable, and
whereas some can infect and cause disease in a broad
range of species, others are restricted to a few related
members of the same genus or family. As in the case of
viruses, cross-protection phenomena have been reported
between viroids. When a plant infected with a mild viroid
strain is challenge-inoculated with a severe strain of the
same viroid or of a closely related viroid, the characteristic
symptoms induced by the latter and its accumulation are
attenuated for a variable time. Some viroids are transmit-
ted through seed and pollen, and at least there is one
well-documented case of aphid transmission. Viroids,
however, spread mainly as a result of agricultural prac-
tices, in particular the use of infected material for vegeta-
tive propagation and of contaminated pruning tools. Within
the plant, viroids, like most viruses, appear to move through
the phloem together with the photosynthetic products
(long-distance movement). Cell-to-cell movement of
viroids seems to occur through plasmodesmata.
Viroid detection was initially performed (and still is in
certain cases) by bioassay in indicator plants, but this
methodology, which is slow and expensive, is being sup-
planted by molecular approaches based on PAGE, hybrid-
ization with radioactive and chemically-labeled cDNA or
cRNA probes, and reverse transcription combined with
polymerase chain reaction (RT-PCR).
4. How viroids replicate and cause
disease
As indicated above, viroids do not appear to code for
any protein [16, 17]. This makes the replication cycle of

947
 R. Flores / C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 324 (2001) 943952

Figure 3. Structural models for viroids. (A) Rod-like secondary structure proposed for members of family Pospiviroidae. The approximate locations
of the five structural domains C (central), P (pathogenic), V (variable), and TL and TR (terminal left and right, respectively) are indicated. The
sequence and location of the TCH (terminal conserved hairpin, present in Hostu- and Cocadviroid genera), TCR (terminal conserved region, present
in Pospi- and Apscaviroid genera, and in CbVd2 and CbVd3) and CCR (central conserved region, here presented that of genus Pospiviroid) are
shown. Arrows denote flanking sequences that, along with the core nucleotides of the upper CCR strand, form imperfect inverted repeats. (B) Left,
quasi rod-like secondary structure proposed for ASBVd, the type member of family Avsunviroidae, with nucleotide residues conserved in all
hammerhead structures shown within boxes with black and white backgrounds for plus and minus polarities, respectively. Right, consensus
hammerhead structure with nucleotide residues conserved in all hammerhead structures shown within boxes with black background and the
arrowhead marking the self-cleavage site. N indicates non-conserved residues and continuous and broken lines denote canonical and
non-canonical base pairs, respectively. The central core is flanked by three helices, I, II and III, that in some cases are closed by short loops.

these molecular parasites extremely host-dependent.
PSTVd [2, 9], the type species of the family Pospiviroidae,
accumulates [18] and replicates [19] in the nucleus,
whereas (ASBVd), the type species of the second viroid
family, Avsunviroidae, accumulates and replicates in the
chloroplast [20, 21]. The available evidence indicates that
other members of both families behave as their corre-
sponding type species [22, 23].
Viroids replicate through RNA intermediates [24] and
are assumed to follow a rolling-circle mechanism that it is
based on the circular nature of the viroid molecule and on
the existence in infected tissues of circular and oligomeric
viroid RNAs of one or both polarities [25]. The latter are
supposed to be the intermediates of the replication circle.
The mechanism may proceed through two alternative
pathways, asymmetric and symmetric, with one and two
operating rolling circles, respectively, and three steps cata-
lyzed by: 1) an RNA polymerase, which after reiterative
RNARNA transcription rounds produces oligomeric
strands, 2) an RNase for cleaving the longer-than-unit
transcripts to unit-length, and 3) an RNA ligase for circu-
larizing the linear monomeric RNAs. Whether the viroid
complementary () oligomers resulting from the first
RNARNA transcription step serve directly as the template
for the second RNARNA transcription, or are previously
cleaved and ligated to the () circular monomer, deter-
mines if the pathway is asymmetric or symmetric, respec-
tively. PSTVd follows the asymmetric pathway [26, 27],
whereas ASBVd follows the alternative symmetric version
[28, 29]. Again, the available data indicate that other
members of both families also behave in the same way as
their type species [23, 30].
Evidence obtained with several viroids of the family
Pospiviroidae indicates that the enzyme catalyzing the
first step of the cycle is the nuclear RNA polymerase II. The
evidence is based on the observation that the low concen

948
 R. Flores / C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 324 (2001) 943952
trations of -amanitin that characteristically inhibit this
enzyme also block viroid replication [19, 31, 32]. In
addition, studies with a monoclonal antibody against a
conserved domain in the largest subunit of RNA poly-
merase II have shown that this subunit co-purifies with the
(+) and () viroid strands [33]. In the case of ASBVd,
experiments using another inhibitor, tagetitoxin, suggest
that a nuclear encoded chloroplastic RNA polymerase
(NEP) is the enzyme involved [34]. Interestingly, initiation
of synthesis of both ASBVd RNAs has been recently
mapped at specific sites in the right terminal loops of their
quasi rod-like structures [35], supporting the notion of a
promoter-driven transcription versus a process initiated at
random, a feasible alternative considering that the circular
nature of the template enables its complete transcription
irrespective of where it starts.
Regarding the RNase catalyzing the second step of the
cycle, cleavage of the oligomeric replicative intermediates
to linear monomeric forms, it was surprising to find firstly
in ASBVd [13] and then in the two other members of
family Avsunviroidae, PLMVd [14] and CChMVd [15],
that it is not an enzyme at all but a hammerhead ribozyme
embedded in both polarity strands (figure 3). Discovery of
ribozymes has sparked a scientific revolution with deeply
rooted implications in several contexts (see below). Within
the family Pospiviroidae, the RNase is generally assumed
to be a host-encoded enzyme, whose specificity is most
probably dictated by a defined conformation of the oligo-
meric viroid RNA [36]. However, the consensus is not
unanimous and other data suggest that cleavage is also
autocatalytic [37]. Finally, the RNA ligase involved in the
third step of the cycle, circularization of linear monomeric
forms, is presumably a host enzyme [36, 38], although it
has been proposed that at least in PLMVd not only cleav-
age but also ligation may be autocatalytic and produce
atypical 25 phosphodiester bonds [39].
Little is known about the mechanisms by means of
which viroids cause disease. Viroid replication does not
necessarily lead to disease and some of these RNAs can
accumulate to relatively high titers without eliciting any
discernible symptom. By contrast, other viroids accumu-
lating even at minimal concentrations are able to trigger
severe diseases. In addition to the macroscopic symptoms,
cell wall malformations, membranous bodies (plasmale-
mmasomes) with distorted organization, and chloroplastic
disturbances have all been observed at the cellular level,
along with changes in the expression of pathogenesis-
related proteins and alterations of hormones and certain
metabolites at the biochemical level. Pathogenicity deter-
minants have been mapped in members of both families to
small regions representing approximately only 1 % of the
total viroid sequence [40, 41], and a single nucleotide
substitution in PSTVd has been shown to expand the host
range of this viroid, enabling it to infect tobacco [42].
However, the primary partners from the host side involved
in these interactions remain unknown.
5. Viroids in an evolutionary context
Although some initial views advanced the existence of
evolutionary links connecting viroids with viruses (and
also with transposable elements and introns) [43], viroids
are now presumed to have an independent and very old
origin, going back to the RNA world postulated to have
existed on Earth before the advent of DNA and proteins.
The need for such a precellular world comes from the
consideration that DNA found in todays cellular world
supports genetic information but is unable to express it,
whereas the opposite occurs with proteins. Since it is
unlikely that DNA and proteins emerged simultaneously
in evolution, there is a compelling need for another older
macromolecule able to fulfill both roles. RNA is an excel-
lent candidate in this respect because it both stores (is the
genetic material of a large number of the existing viruses)
and expresses (an increasing number of catalytic RNAs
with ribozyme activity have been described along the
recent years) genetic information. The finding that some
viroids, particularly those forming family Avsunviroidae,
contain hammerhead ribozymes, together with other viroid
characteristics such as circular structure, small size and
high G+C content, has led to the proposal that they may be
living fossils of the RNA world that, after the appearance
of cellular organisms, would have adopted an intracellular
mode of existence [44]. At this stage, mutations and recom-
bination events between viroids co-infecting the same cell
must have certainly contributed to their divergence.
Are viroids restricted to the plant kingdom? In strict
terms the answer must be yes, for the moment. There are
hints indicating that viroids might well invade other cell
types. The fact that some viroids replicate and accumulate
in the nucleus (a characteristic organelle of eukaryotic
cells) and others in the chloroplast (an organelle evolved
by symbiosis from cyanobacteria), speaks clearly of the
versatility of these parasitic RNAs and suggests the possi-
bility that viroids may exist in present-day cyanobacteria.
On the other hand, the RNA of human hepatitis delta virus
(HDV) shares some features with viroids: it is circular,
adopts a rod-like secondary structure and replicates
through a rolling-circle mechanism with the cleavage step
catalyzed by a special class of ribozymes [45]. However,
in contrast to viroids, HDV RNA is considerably larger
(1 700 nt), has the ability to code for a protein in its
antigenomic polarity and is functionally dependent on
another virus, hepatitis B virus. Detailed analysis of the
HDV RNA structure has revealed the existence of a viroid-
like domain (containing the ribozymes) fused to a protein
coding domain [46], and recent data indicate the exist-
ence in uninfected liver of a protein that can influence
HDV replication and that appears to be related to the HDV
protein [47]. These results suggest the intriguing possibility
that HDV RNA may have arisen in evolution by the
capture of a messenger RNA encoding this protein by a
self-replicating viroid-like RNA. Therefore, it would not be
surprising at all if viroids are finally discovered in bacteria,
fungi or even in animals.
949
 R. Flores / C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 324 (2001) 943952
6. Future avenues for viroid research
From a phytopathological perspective, most of the
viroids causing diseases in major economic crops have
been presumably identified. However, there may still be
quite a large number of viroids infecting wild plants,
which do not usually exhibit symptoms and may thus act
as reservoirs from where viroids may eventually jump to
cultivated species and incite new diseases. Modern agri-
culture practices, characterized by extensive plantations
of genetically uniform species and intensive movement of
plant material worldwide, may have facilitated jumps of
this type. Consistent with this view, all viroid diseases have
been recorded in the XX century [48].
On the other hand, certain effects induced by viroids
can be manipulated to improve the agronomic perfor-
mance of specific plants. This appears to be the case of the
dwarfing effects of some citrus viroids that can be used for
high density citrus cultivation that out-yield conventional
groves without any apparently deleterious side conse-
quences (Hutton et al., [49]). Also, in a biotechnological
context, hammerhead ribozymes, which in their natural
habitat act in cis cleaving the RNAs in which they are
contained, can be manipulated to act in trans as restric-
tion RNases to degrade specific RNAs, an approach with
an enormous potential influence in medicine and industry.
From a more academic standpoint, many issues still
remain unresolved. These include a detailed characteriza-
tion of host proteins assisting the three steps of the repli-
cation cycle and those involved in viroid movement, the
molecular determinants targeting some viroids to the
nucleus and others to the choroplast, and the initial viroid-
host interaction that triggers a signal transduction cas-
cade ultimately leading to the macroscopic symptoms.
Detailed analysis of viroidhost interactions should fur-
nish hints on how the cell metabolism is subverted by
these unique pathogens and, perhaps, also on how the cell
functions under normal physiological conditions. More-
over, as already indicated, studies on these minimal genetic
systems may provide clues to early steps of the appearance
and evolution of life on our planet. As Pasteur put it:
Limportance de linfiniment petit est infiniment grande
(The importance of the infinitely small is infinitely large),
an idea also advanced by a great Roman naturalist: Natura
numquam magis est tota quam in minimis (In no other
place is nature as a whole as in its smallest creatures)
(Plinio, dead in 79 B.C., victim of his own scientific
curiosity while observing Vesuvius eruption).
7. Compte-rendu de la sance commune
Acadmie des Sciences / Acadmie
dAgriculture de France, 5. 2. 2001
Expos de M. Ricardo Flores, Un RNA nu, spcifique de
plante et dix fois plus petit que le plus petit RNA de virus :
virode
950
Question pose par M. Claude Hlne (Acadmie des
Sciences)
Quelle est la nuclase cellulaire qui coupe le produit de
rplication de lARN du virode en fragments correspon-
dant exactement lARN du virode ?
. Rponse
In members of the Avsunviroidae family the nuclease is
not an enzyme but a hammerhead ribozyme embedded in
the RNAs of both polarities. There is strong evidence
supporting that these ribozymes are not only active in vitro
but also in vivo. Although in vivo they may be assisted by
host proteins that may considerably enhance their effi-
ciency, the chemistry of the catalytic mechanism is almost
certainly RNA-based. In members of the Pospiviroidae
family, the cleavage reaction is generally assumed to be
catalysed by an unidentified host RNase. In fact, a nuclear
extract from potato cells can process in vitro linear oligo-
meric PSTVd RNAs into the infectious monomeric circular
RNA. However, a heterologous nuclease such as the fun-
gal RNase T1 can perform the same function. Although
these results could be regarded as supporting that a host
RNase catalyses both steps, the alternative view of an
RNase producing linear monomers with 5-hydroxyl and
2,3-cyclic phosphate termini, which are then circularized
by an RNA ligase, seems more likely given that an RNA
ligase mediates the in vitro circularization of naturally
occurring PSTVd linear monomers. The specifity of the
cleavage reaction or, in other words, the location of the
processing/ligation site, is most probably dictated by a
defined conformation of the oligomeric viroid RNA. How-
ever, there is still much controversy surrounding these
issues: i) some data indicate that the exact
processing/ligation site is located in the upper strand of the
CCR (central conserved region) whereas other data point
to the existence of alternative sites, ii) it has been proposed
that the cleavage step in family Pospiviroidae may also be
autocatalytic and mediated through non-hammerhead
ribozymes, and iii) at least in one member of the family
Avsunviroidae it has been advanced that not only cleav-
age, but also ligation may be autocatalytic and produce
atypical 2-5 phosphodiester bonds.
Quel est (ou sont) le(s) mcanisme(s) de la pathogni-
cit des virodes?
Question pose par Mme Nicole Le Douarin (Acadmie
des Sciences)
. Rponse de M. Flores
Mechanisms underlying viroid-host cell interactions,
more specifically those leading the onset of symptoms, are
poorly understood at present. Most likely they are diverse
in nature because members of family Pospiviroidae repli-
cate and accumulate in the nucleus whereas these pro-
cesses occur in the chloroplast in members of family
Avsunviroidae. From the viroid side, minimal sequence
changes affecting approximately 1% of the sequence are
sufficient to change a severe into a latent strain. Phenom
 R. Flores / C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 324 (2001) 943952
ena of this kind have been observed with representative
members of both families and have permitted to define the
virulence-modulating region in PSTVd and a pathoge-
nicity determinant in CChMVd. The effects of these
changes are not the result of an impaired replication
because viroid titers are essentially the same. The hypoth-
esis that appears more reliable is that these changes affect
the interaction of the viroid with a host factor. Regarding
the nature of this factor it has been proposed that it could
be a small cellular RNA or a protein. Although the first
References
[1] Bos L., The embryonic beginning of virology: unbiased thinking
and dogmatic stagnation, Arch. Virol. 140 (1995) 613619.
[2] Diener T.O., Potato spindle tuber virus IV. A replicating low
molecular weight RNA, Virology 45 (1971) 411428.
[3] Diener T.O., Viroids and the nature of viroid diseases, Arch. Virol.
(suppl.) 15 (1999) 203220.
[4] Diener T.O., Potato spindle tuber viroid VIII. Correlation of infec-
tivity with a UV-absorbing component and thermal denaturation proper-
ties of the RNA, Virology 50 (1972) 606609.
[5] Sogo J.M., Koller T., Diener T.O., Potato spindle tuber viroid X.
Visualization and size determination by electron microscopy, Virology 55
(1973) 7080.
[6] Snger H.L., Klotz G., Riesner D., Gross H.J., Kleinschmidt A.,
Viroids are single-stranded covalently closed circular RNA molecules
existing as highly base-paired rod-like structures, Proc. Natl. Acad. Sci.
USA 73 (1976) 38523856.
[7] Semancik J.S., Weathers L.G., Exocortis disease: Evidence for a new
species of infectious low molecular weight RNA in plants, Nat. New
Biol. 237 (1972) 242244.
[8] Flores R., Randles J.W., Bar-Joseph M., Diener T.O., Subviral agents:
Viroids, in: Van Regenmortel M.H.V., Fauquet C.M., Bishop D.H.L.,
Carstens E.B., Estes M.K., Lemon S.M., Maniloff J., Mayo M.A.,
McGeoch D.J., Pringle C.R., Wickner R.B. (Eds.), Virus Taxonomy, Sev-
enth Report of the International Committee on Taxonomy of Viruses,
Academic Press, San Diego, 2000, pp. 10091024.
[9] Gross H.J., Domdey H., Lossow C., Jank P., Raba M., Alberty H.,
Snger H.L., Nucleotide sequence and secondary structure of potato
spindle tuber viroid, Nature 273 (1978) 203208.
[10] Keese P., Symons R.H., Domains in viroids: Evidence of intermo-
lecular RNA rearrangements and their contribution to viroid evolution,
Proc. Natl. Acad. Sci. USA 82 (1985) 45824586.
[11] Sano T., Candresse T., Hammond R.W., Diener T.O., Owens R.A.,
Identification of multiple structural domains regulating viroid pathogenic-
ity, Proc. Natl. Acad. Sci. USA 89 (1992) 1010410108.
[12] Symons R.H., Avocado sunblotch viroid: Primary sequence and
proposed secondary structure, Nucleic Acids Res. 9 (1981) 65276537.
[13] Hutchins C., Rathjen P.D., Forster A.C., Symons R.H., Self-cleavage
of plus and minus RNA transcripts of avocado sunblotch viroid, Nucleic
Acids Res 14 (1986) 36273640.
[14] Hernndez C., Flores R., Plus and minus RNAs of peach latent
mosaic viroid self-cleave in vitro via hammerhead structures, Proc. Natl.
Acad. Sci. USA 89 (1992) 37113715.
[15] Navarro B., Flores R., Chrysanthemum chlorotic mottle viroid:
Unusual structural properties of a subgroup of viroids with hammerhead
ribozymes, Proc. Natl. Acad. Sci. USA 94 (1997) 1126211267.
[16] Davies J.W., Kaesberg P., Diener T.O., Potato spindle tuber viroid.
XII. An investigation of viroid RNA as a messenger for protein synthesis,
Virology 61 (1974) 281286.
[17] Hall T.C., Wepprich R.K., Davies J.W., Weathers L.G., Seman-
cik J.S., Functional distinctions between the ribonucleic acids from citrus
exocortis viroid and plant viruses: Cell-free translation and aminoacyla-
tion reactions, Virology 61 (1974) 486492.
[18] Harders J., Lukacs N., Robert-Nicoud M., Jovin J.M., Riesner D.,
Imaging of viroids in nuclei from tomato leaf tissue by in situ hybridization
and confocal laser scanning microscopy, EMBO J. 8 (1989) 39413949.
alternative can not be excluded, the second one appears
more likely because viroids must necessarily interact with
host proteins for their replication and, probably, for their
movement and accumulation. It is conceivable to assume
that as a result of one or more of these interactions, a
protein could be diverted from its normal physiological
role and that this could lead to a primary alteration that
following a signal transduction pathway cause ultimately
the onset of symptoms. The nature of this protein remains
to be determined.
[19] Schindler I.M., Mhlbach H.P., Involvement of nuclear DNA-
dependent RNA polymerases in potato spindle tuber viroid replication: A
reevaluation, Plant Sci. 84 (1992) 221229.
[20] Bonfiglioli R.G., McFadden G.I., Symons R.H., In situ hybridiza-
tion localizes avocado sunblotch viroid on chloroplast thylakoid mem-
branes and coconut cadang cadang viroid in the nucleus, Plant J. 6 (1994)
99103.
[21] Lima M.I., Fonseca M.E.N., Flores R., Kitajima E.W., Detection of
avocado sunblotch viroid in chloroplasts of avocado leaves by in situ
hybridization, Arch. Virol. 138 (1994) 385390.
[22] Bonfiglioli R.G., Webb D.R., Symons R.H., Tissue and intra-
cellular distribution of coconut cadang cadang viroid and citrus exocortis
viroid determined by in situ hybridization and confocal laser scanning
and transmission electron microscopy, Plant J. 9 (1996) 457465.
[23] Bussire F., Lehoux J., Thompson D.A., Skrzeczkowsk I.L.J., Per-
reault J.P., Subcellular localization and rolling circle replication of peach
latent mosaic viroid: Hallmarks of group A viroids, J. Virol 73 (1999)
63536360.
[24] Grill L.K., Semancik J.S., RNA sequences complementary to citrus
exocortis viroid in nucleic acid preparations from infected Gynura auran-
tiaca, Proc. Natl. Acad. Sci. USA 75 (1978) 896900.
[25] Branch A.D., Robertson H.D., A replication cycle for viroids and
other small infectious RNAs, Science 223 (1984) 450455.
[26] Branch A.D., Benenfeld B.J., Robertson H.D., Evidence for a
single rolling circle in the replication of potato spindle tuber viroid, Proc.
Natl. Acad. Sci. USA 85 (1988) 91289132.
[27] Feldstein P.A., Hu Y., Owens R.A., Precisely full length, circular-
izable, complementary RNA: An infectious form of potato spindle tuber
viroid, Proc. Natl. Acad. Sci. USA 95 (1998) 65606565.
[28] Hutchins C.J., Keese P., Visvader J.E., Rathjen P.D., McInnes J.L.,
Symons R.H., Comparison of multimeric plus and minus forms of viroids
and virusoids, Plant Mol. Biol. 4 (1985) 293304.
[29] Dars J.A., Marcos J.F., Hernndez C., Flores R., Replication of
avocado sunblotch viroid: Evidence for a symmetric pathway with two
rolling circles and hammerhead ribozyme processing, Proc. Natl. Acad.
Sci. USA 91 (1994) 1281312817.
[30] Ishikawa M., MeshI T., Ohno T., Okada Y., Sano T., Ueda I.,
Shikata E., A revised replication cycle for viroids: The role of longer than
unit RNA in viroid replication, Mol. Gen. Genet. 196 (1984) 421428.
[31] Mhlbach H.P., Snger H.L., Viroid replication is inhibited by
-amanitin, Nature 278 (1979) 185188.
[32] Flores R., Semancik J.S., Properties of a cell-free system for syn-
thesis of citrus exocortis viroid, Proc. Natl. Acad. Sci. USA 79 (1982)
62856288.
[33] Warrilow D., Symons R.H., Citrus exocortis viroid RNA is associ-
ated with the largest subunit of RNA polymerase II in tomato in vivo, Arch.
Virol. 144 (1999) 23672375.
[34] Navarro J.A., Vera A., Flores R., A chloroplastic RNA polymerase
resistant to tagetitoxin is involved in replication of avocado sunblotch
viroid, Virology 268 (2000) 218225.
[35] Navarro J.A., Flores R., Characterization of the initiation sites of
both polarity strands of a viroid RNA reveals a motif conserved in
sequence and structure, EMBO J. 19 (2000) 26622670.
[36] Baumstark T., Schrder A.R.W., Riesner D., Viroid processing:
switch from cleavage to ligation is driven by a change from a tetraloop to
a loop E conformation, EMBO J. 16 (1997) 599610.
[37] Liu Y.H., Symons R.H., Specific RNA self-cleavage in coconut
cadang cadang viroid: Potential for a role in rolling circle replication, RNA
4 (1998) 418429.
951
 R. Flores / C.R. Acad. Sci. Paris, Sciences de la vie / Life Sciences 324 (2001) 943952
[38] Tsagris M., Tabler M., Mhlbach H.P., Snger H.L., Linear oligo-
meric potato spindle tuber viroid (PSTV) RNAs are accurately processed in
vitro to the monomeric circular viroid proper when incubated with a
nuclear extract from healthy potato cells, EMBO J. 6 (1987) 21732183.
[39] Ct F., Lvesque D., Perreault J.P., Natural 2,5-phosphodiester
bonds found at the ligation sites of peach latent mosaic viroid, J. Virol. 75
(2001) 1925.
[40] Gross H.J., Liebl U., Alberty H., Krupp G., Domdey H., Ramm K.,
Snger H.L., A severe and a mild potato spindle tuber viroid isolate differ
in three nucleotide exchanges only, Biosci. Rep. 1 (1981) 235241.
[41] De la Pea M., Navarro B., Flores R., 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
(1999) 99609965.
[42] Wassenegger M., Spieker R.L., Thalmeir S., Gast F.U., Riedel L.,
Snger H.L., A single nucleotide substitution converts potato spindle tuber
viroid (PSTVd) from a noninfectious to an infectious RNA for Nicotiana
tabacum, Virology 226 (1996) 191197.
952
[43] Diener T.O., Origin and evolution of viroids and viroid-like satel-
lite RNAs, Virus Genes 11 (1996) 119131.
[44] Diener T.O., Circular RNAs: Relics of precellular evolution? Proc.
Natl. Acad. Sci. USA 86 (1989) 93709374.
[45] Taylor J.M., Human hepatitis delta virus: An agent with similarities
to certain satellite RNAs of plants, in: Vogt P.K., Jackson A.O. (Eds.),
Current Topics in Microbiology and Immunology, vol. 329, Satellites and
Defective Viral RNAs, Springer, Berlin, 1999, pp. 107122.
[46] Robertson H.D., How did replicating and coding RNAs first get
together? Science 274 (1996) 6667.
[47] Brazas R., Ganem D., A cellular homolog of hepatitis delta
antigen: Implications for viral replication and evolution, Science 274
(1996) 9094.
[48] Diener T.O., Viroids and Viroid Diseases, Wiley, New York, 1979.
[49] Hutton R.J., Broadbent P., Bevington K.B., Viroid dwarfing for high
density citrus plantings, Hortic. Rev. 44 (2000) 277317.