The Plant Cell, Vol. 15, 1360–1374, June 2003, www.plantcell.org © 2003 American Society of Plant Biologists

Inhibition of Cell Growth and Shoot Development by a Specific Nucleotide Sequence in a Noncoding Viroid RNA

Yijun Qi and Biao Ding

1

Department of Plant Biology and Plant Biotechnology Center, Ohio State University, Columbus, Ohio 43210

Viroids are small noncoding and infectious RNAs that replicate autonomously and move systemically throughout an in-fected plant. The RNAs of the family Pospiviroidae contain a central conserved region (CCR) that has long been thought tobe involved in replication. Here, we report that the CCR of

Potato spindle tuber viroid

(PSTVd) also plays a role in pathoge-nicity. A U257A change in the CCR converted the intermediate strain PSTVd

Int

to a lethal strain that caused severe growthstunting and premature death of infected plants. PSTVd with nucleotide U257 changed to C or G did not cause such symp-toms. The pathogenic effect of the U257A substitution was abolished by a C259U substitution in the same RNA. Analyses ofthe pathogenic effects of the U257A substitution in three other PSTVd variants established A257 as a new pathogenicity de-terminant that functions independently and synergistically with the classic pathogenicity domain. The U257A substitutiondid not alter PSTVd secondary structure, replication levels, or tissue tropism. The stunted growth of PSTVd

Int

U257A-infectedtomato plants resulted from restricted cell expansion but not cell division or differentiation. This was correlated positivelywith the downregulated expression of an expansin gene,

LeExp2

. Our results demonstrate that specific nucleotides in anoncoding, pathogenic RNA have a profound effect in altering distinct cellular responses, which then lead to well-definedalterations in plant growth and developmental patterns. The feasibility of correlating viroid RNA sequence/structure withthe altered expression of specific host genes, cellular processes, and developmental patterns makes viroid infection a valu-able system in which to investigate host factors for symptom expression and perhaps also to characterize the mechanismsof RNA regulation of gene expression in plants.

INTRODUCTION

Development of disease symptoms in a host plant when in-fected by a particular pathogen is the result of complex patho-gen–host interactions. The pathogen usually has pathogenicitydeterminants that elicit the types and degrees of severity ofhost symptoms. In the host, altered cellular functions lead tochanges in physiology and/or development that exhibit as dis-ease symptoms. Thus, understanding the cellular and molecu-lar mechanisms of disease formation will not only provide thebasis for the development of rational approaches to combatpathogen infection but also will provide insights about the ba-sic cellular processes that underlie normal plant growth and de-velopment (Kasschau et al., 2003).

Viroid infection provides a simple experimental system inwhich to study direct interactions between a pathogenic RNAgenome and the host. A viroid genome consists of a single-stranded, covalently closed circular, noncoding, and nonen-capsidated RNA. Without encoding capacity, the viroid RNAgenome and its replication intermediate interact directly withhost components for nearly all aspects of the infection process,including replication, intracellular movement, intercellular move-ment, systemic movement, and pathogenicity (Hull, 2002).

The pleiotropic functions of the viroid RNA genome are re-

markable given its small size, structural simplicity, and noncod-ing capacity. Viroid RNAs range from 246 to 399 nucleotides insize (Flores et al., 1997). Viroids in the Pospiviroidae family as-sume a rod-shaped secondary structure in their native state,and some viroids in the Avsunviroidae family have branchedstructures (Flores et al., 2000). A rod-shaped viroid RNA, asshown in Figure 1, contains five broad structural domains: (1) acentral conserved region (CCR) involved in replication; (2) a path-ogenicity domain implicated in symptom expression; (3) a vari-able domain; and (4) right- and (5) left-terminal domains (Keeseand Symons, 1985). During replication, metastable structuressuch as hairpins may be formed through structural rearrange-ment (Loss et al., 1991; Owens et al., 1991; Qu et al., 1993;Schröder and Riesner, 2002). Local tertiary structures also mayexist (Branch et al., 1985; Gast et al., 1996). A viroid is essen-tially an RNA mosaic with tightly packed and perhaps overlap-ping structural and functional motifs.

Viroid-infected plants can be symptomless or develop symp-toms that range from mild to lethal, depending on viroid strainsand host species (Schnölzer et al., 1985; Owens, 1990; Owenset al., 1991, 1995, 1996; Hammond, 1992; de la Peña et al.,1999; kori et al., 2001). Because viroid pathogenesis is aconsequence of direct RNA–cellular factor interactions, viroidsare considered to belong to a group of noncoding RNAs thatregulate cellular functions through means other than by encod-ing proteins for specific functions (Storz, 2002). In this context,elucidating the molecular mechanisms of viroid disease mayhelp us understand the mechanisms of the RNA regulation ofcellular processes.

S c

1

To whom correspondence should be addressed. E-mail ding.35@osu.edu; fax 614-292-5379.Article, publication date, and citation information can be found atwww.plantcell.org/cgi/doi/10.1105/tpc.011585.

RNA Inhibition of Cell Growth 1361

Potato spindle tuber viroid

(PSTVd) is the type species of thePospiviroidae family (Flores et al., 2000). Sequence variationsthat contribute to different degrees of symptom severity havebeen mapped to the pathogenicity domain (Schnölzer et al.,1985; Owens et al., 1991, 1995, 1996; Hammond, 1992). Moststudies have been limited to analyses of viroid structure andgeneral plant symptoms (Schnölzer et al., 1985; Owens, 1990;Owens et al., 1991, 1995, 1996; Hammond, 1992; Schmitz andRiesner, 1998). Changes in the global gene expression patternsof infected plants also have been described (Itaya et al., 2002).In general, however, we have little knowledge of how a specificPSTVd sequence or structure can evoke distinct changes inhost gene expression that lead to alterations in specific cellularprocesses and the development of particular symptoms.

We have taken a comprehensive approach that includes mo-lecular, cellular, biophysical, and whole-plant analyses to inves-tigate the mechanisms of viroid pathogenicity using PSTVd in-fection of tomato as an experimental system. Here, we reportthat a specific nucleotide change in CCR, a region conserved inall viroids of the family Pospiviroidae and long thought to be in-volved in replication (Keese and Symons, 1985) and host rangedetermination (Wassenegger et al., 1996), confers a novel lethalsymptom on the infected tomato plants. Underlying this symp-tom is inhibited cell growth and shoot development, marked bythe repressed expression of a tomato expansin gene implicatedin cell growth. The biological implications of these results arediscussed.

RESULTS

The Nucleotide Substitution U257A Converted PSTVd

Int

to a Lethal Strain in Tomato

The predicted secondary structure of the intermediate PSTVdstrain (PSTVd

Int

) with nucleotide sequences (Gross et al., 1978)is shown in Figure 1A. We showed previously that two indepen-dent mutations in loop E of CCR, C259U and U257A, convertedPSTVd

Int

to tobacco infectious variants PSTVd

Int

C259U andPSTVd

Int

U257A, respectively (Figure 1B) (Qi and Ding, 2002;Zhu et al., 2002). We were interested in knowing whether thesemutations would alter infectivity in tomato, a convenient experi-mental host for the study of viroid symptoms. We inoculatedtomato seedlings at the cotyledon stage (6 days old, before thefirst leaf was visible) with in vitro transcripts of PSTVd

Int

and itstwo variants. Water was used as the inoculum in mock inocula-tion. As shown in Figure 2A, PSTVd

Int

C259U infection causedsymptoms similar to those caused by PSTVd

Int

. Strikingly, plantsinfected with PSTVd

Int

U257A showed severe growth stuntingand relatively small leaves (Figure 2A). These plants displayed acharacteristic flat appearance on the top of the shoot, with alllateral organs (leaves) at similar vertical levels (Figure 2B). Wedesignate this phenotype the “flat-top” symptom. Furthermore,the leaves showed yellowing and necrosis. Sequence analysesindicated that PSTVd progeny maintained the mutated se-quences in the infected plants (data not shown).

Figure 1. Secondary Structure of PSTVd.

(A) The viroid structure and nucleotide numbering are based on Gross et al. (1978). The viroid contains left terminal (TL; 1 to 46/315 to 359), pathoge-nicity (47 to 73/286 to 314), central conserved (74 to 120/240 to 285), variable (121 to 148/212 to 239), and right terminal (TR; 149 to 211) domains asproposed by Keese and Symons (1985). The structure of loop E in the central conserved domain is enlarged. Stars indicate noncanonical base-pair-ing, and the twisted line connecting G98 and U260 represents the loop E–specific UV cross-linking (Branch et al., 1985; Baumstark et al., 1997).(B) Selected portions of PSTVdInt, PSTVdIntC259U, and PSTVdIntU257A containing nucleotides of loop E. The nucleotide substitutions C259U andU257A that give rise to the latter two variants are indicated by arrows. (Figure adapted from Zhu et al., 2002.)

1362 The Plant Cell

Figure 2. Symptoms Caused by Infection of Rutgers Tomato with PSTVd Variants.

Cotyledons of 6-day-old seedlings were inoculated with 100 ng/�L PSTVd transcripts. Photographs were taken and lengths of internodes were mea-sured at 6 weeks after inoculation.(A) PSTVdIntC259U and PSTVdIntU257C cause symptoms similar to those caused by PSTVdInt. PSTVdIntU257A causes severe growth stunting, flattop, and premature plant death. PSTVdIntU257G causes growth stunting intermediate between that caused by PSTVdIntC259U and PSTVdIntU257A.(B) Closer view of the PSTVdIntU257A-infected plant showing leaf necrosis and the flat-top appearance of the shoot tip.(C) Quantitative analyses of the stunted growth of PSTVd-infected tomato plants at 6 weeks after inoculation. The lengths of individual color bars rep-resent the lengths of successive internodes, with the first internode represented at bottom. Each value is the average of measurements from six plantsinfected with each PSTVd variant.(D) The double mutant PSTVdIntC259U/U257A causes a symptom similar to that caused by PSTVdIntC259U or PSTVdInt.(E) to (G) The U257A substitution in parental PSTVdMild (E), PSTVdKF440-2 (F), and PSTVdRG1 (G) causes the flat-top symptom in each case. Note thenormal “pyramid-shaped” appearance of shoot tips of plants infected with each of the parental viroids.

RNA Inhibition of Cell Growth 1363

The heights of the plants and the lengths of successive inter-nodes were measured from six plants infected with thesePSTVd variants at 6 weeks after inoculation. The internodeswere numbered sequentially, with the first one being immedi-ately above the cotyledons. As shown in Figure 2C, the lengthsof the first and second internodes were similar among plants in-fected with PSTVd

Int

, PSTVd

Int

C259U, or PSTVd

Int

U257A. Start-ing from the third internode, the PSTVd

Int

U257A-infected plantsshowed shortened length compared with plants infected withPSTVd

Int

or PSTVd

Int

C259U. The magnitude of shorteningincreased with subsequent internodes (Figure 2C). Thus,PSTVd

Int

U257A-infected plants developed the flat-top symp-tom as a result of increased shortening or stunted growth ofsuccessive internodes.

At 7 weeks after inoculation, the leaves of PSTVd

Int

U257A-infected tomato plants started to die. These plants did notproduce flowers. At the same time, plants infected withPSTVd

Int

C259U or PSTVd

Int

had produced flowers, althoughthey were delayed compared with flower production in mock-inoculated plants.

The U257A Substitution in PSTVd

Int

Was Required Specifically for the Development of the Flat-Top Symptom

To determine whether the absence of U or the presence of A atposition 257 was critical for PSTVd

Int

U257A to cause the flat-top symptom in tomato, we inoculated tomato plants withPSTVd

Int

-derived variants that have U257 replaced with nucle-otide C or G by site-directed mutagenesis (Qi and Ding, 2002).As shown in Figure 2A, PSTVd

Int

U257C caused a symptomsimilar to that caused by PSTVd

Int

. On the other hand,PSTVd

Int

U257G caused growth stunting intermediate betweenthat caused by PSTVd

Int

U257A and PSTVd

Int

C259U (Figures 2Aand 2C). These results indicate that nucleotide identity at posi-tion 257 plays a role in symptom expression and that A257 isrequired specifically for the development of the severe stuntingand flat-top symptoms.

The U257A substitution could have caused changes in thesecondary or higher structure of PSTVd to evoke new interac-tions with host factors to produce the flat-top symptom. Alter-natively, the modified CCR could confer the pathogenic effectat the nucleotide sequence level. We showed previously thatthe computed PSTVd secondary structure was not altered bythe U257A substitution (Qi and Ding, 2002; Zhu et al., 2002). Toverify the computed data, we used temperature gradient gelelectrophoresis (TGGE) (Figure 3A) to analyze the denaturationprofiles of PSTVd

Int

U257A and PSTVd

Int

. A comparison of thedenaturation profiles of different PSTVd RNAs on the sameTGGE gel would reveal their structural similarities or differences(Rosenbaum and Riesner, 1987; Riesner, 1991). Our resultsshowed that PSTVd

Int

and PSTVd

Int

U257A exhibited indistin-guishable denaturation profiles in TGGE (Figure 3B). As a con-trol, PSTVd

Int

and variant PSTVd

Int

A135G exhibited the ex-pected differences in thermal stability (Figure 3C) (Owens et al.,1996). Similar differences were observed between PSTVd

Int

U257Aand PSTVd

Int

A135G (Figure 3D), further confirming that PSTVd

Int

and PSTVd

Int

U257A share the same thermal stability.These structural analyses indicate that the U257A substitu-

tion does not cause significant structural changes in PSTVd invitro. Whether the same holds true in vivo remains to be deter-mined. Nonetheless, our data suggest that a local change inthe CCR sequence most likely was responsible for the induc-tion of the flat-top symptom. The specificity of the pathogeniceffect of A257 in the CCR was further established in the experi-ments described below.

The Symptoms Caused by the U257A Substitution Were Abolished by the C259U Substitution in a DoublePSTVd Mutant

Given that the U257A substitution caused severe growth-stunt-ing and flat-top symptoms and the C259U substitution did not,we tested the effect of the presence of both of these mutationson symptom expression. We created a double mutant,PSTVd

Int

C259U/U257A, that contains both mutations in thesame viroid RNA (Qi and Ding, 2002). Surprisingly, the dou-ble mutant caused symptoms similar to those caused byPSTVd

Int

C259U (Figures 2C and 2D). Thus, host interactionsmediated by U259 appear to dominate interactions mediatedby A257. Alternatively, the presence of U259 may abolish theinteractions of A257 with host factors. Regardless of the spe-cific mechanisms, these data provide additional evidence for

Figure 3. Thermal Stability of PSTVd Variants Determined by TGGE.

(A) Scheme of TGGE. Mixtures of approximately equal amounts of 32P-labeled PSTVd transcripts were analyzed on 5% polyacrylamide gels(acrylamide:bisacrylamide [30:1], 0.2 � TBE, and 5 mM NaCl) with a 25to 60�C temperature gradient. After electrophoresis, the gels were fixedand subjected to autoradiography.(B) Denaturation profiles of PSTVdInt and PSTVdIntU257A, which are in-distinguishable and appear as a single band (arrow). C and L denote cir-cular and linear PSTVd, respectively.(C) and (D) The denaturation profile of PSTVdIntA135G is different fromthose of PSTVdInt (C) and PSTVdIntU257A (D).

1364 The Plant Cell

the specificity of A257-conferred host interactions for the ex-pression of the flat-top symptom. Furthermore, these datademonstrate that the effect of A257 can be modulated by otherspecific nucleotides in the CCR.

The CCR and the Pathogenicity Domain Function Independently and Synergistically in Symptom Expression

To test the general importance of A257 in pathogenicity, we re-placed nucleotide U257 with A in three other PSTVd variants,PSTVd

Mild

, PSTVd

KF440-2

, and PSTVd

RG1

, by site-directed mu-tagenesis. These variants differ in nucleotide sequences in thevirulence-modulating region of the classic pathogenicity do-main (Figure 4). PSTVd

Mild

causes mild symptoms (Schnölzer etal., 1985), whereas PSTVd

KF440-2

(Schnölzer et al., 1985) andPSTVd

RG1

(Gruner et al., 1996) cause lethal symptoms in in-fected tomato.

Tomato plants infected with the new PSTVd variants(PSTVd

Mild

U257A, PSTVd

KF440-2

U257A, and PSTVd

RG1

U257A)also developed the characteristic flat-top symptom in additionto further growth stunting in each case (Figures 2C to 2G).These data indicate that A257 plays a specific role in conferringthe flat-top symptom, regardless of the parental PSTVd back-ground. Furthermore, the classic pathogenicity domain canmodulate the pathogenic effects of A257. Sequence variationsin this domain allow infected plants to develop different severitylevels of the final symptoms. Therefore, the classic pathogenic-

ity domain and the CCR appear to function independently andsynergistically in symptom expression.

The U257A Substitution Did Not Enhance PSTVd Accumulation in Infected Tomato

Increased severity of a plant disease could be the result of en-hanced accumulation of a pathogen. Both the U257A andC259U substitutions in PSTVd

Int

enhance replication in tobaccocells (Qi and Ding, 2002), likely accounting for the increased in-fectivity in tobacco plants (Zhu et al., 2002). Therefore, weasked whether these nucleotide substitutions also could en-hance PSTVd accumulation in tomato and whether U257A par-ticularly would enhance accumulation in tomato to account forthe development of the flat-top symptom.

To analyze the replication capacity of PSTVd

Int

C259U andPSTVd

Int

U257A compared with that of PSTVd

Int

at the whole-plant and cellular levels, we inoculated tomato plants as well asprotoplasts with in vitro transcripts of these PSTVd variants.Total RNAs were extracted from infected plants at 1-week in-tervals for up to 6 weeks and from infected protoplasts at 3days after inoculation. Accumulation levels of PSTVd were de-termined by RNA gel blot analysis. Such analyses showed thatthese PSTVd variants accumulated to similar levels in infectedwhole plants (Figures 5A and 5B) and infected protoplasts (Fig-ure 5C). Thus, in contrast to the situation in tobacco, the U257Aand C259U substitutions did not enhance PSTVd accumulationat either the cellular or the whole-plant level in tomato. There-fore, the flat-top symptom caused by PSTVd

Int

U257A infectionwas not correlated with viroid accumulation levels.

The U257A Substitution Did Not Alter Tissue Tropism of PSTVd in Infected Tomato

It was possible that PSTVd

Int

U257A invaded tomato cells thatwere not invaded normally by PSTVd

Int

, causing the flat-topsymptom in the infected plants. Given the central role of theshoot apical meristem (SAM) in development (Sussex, 1989)and the observation that PSTVd

Int

did not invade the tomatoSAM (Zhu et al., 2001), we asked whether PSTVd

Int

U257Awould invade the SAM to interfere with its normal function,leading to perturbed development. To address this question,we performed in situ hybridization experiments using ribo-probes specific for the plus-strand PSTVd. We examined SAMsof tomato plants infected with PSTVd

Int

and PSTVd

Int

U257A.Consistent with previous observations (Zhu et al., 2001), PSTVd

Int

was absent from the SAM (Figures 6A and 6B). PSTVd

Int

U257Aalso was absent from the SAM (Figures 6C and 6D).

We also extended the analyses of PSTVd localization toleaves and stems of infected plants. The two PSTVd variantswere present in all tissues of leaves (Figures 7A to 7D) andstems (Figures 7E to 7H). Thus, the U257A substitution did notalter the cellular localization patterns of PSTVd in tomato.

Because the U257A substitution did not alter the accumula-tion levels or tissue tropism of PSTVd, we attributed its patho-genic effect to specific molecular interactions with a host factor(s).Such interactions resulted in inhibited shoot development toaccount for the flat-top symptom. To elucidate the basis of the

Figure 4. Secondary Structures of the Virulence-Modulating Region ofthe Classic Pathogenicity Domain of PSTVd Variants.

The secondary structures were computed at 25�C by mfold version 3.0(Water et al., 1994). They are identical to those reported in other studies(Schnölzer et al., 1985; Owens et al., 1996). The marked nucleotide in-sertion (�), deletion (�), or substitution in PSTVdMild, PSTVdKF440-2, andPSTVdRG1 are deviations from the nucleotide sequence of PSTVdInt.

RNA Inhibition of Cell Growth 1365

molecular interactions that led to the inhibited shoot development,we performed experiments to identify key cellular processes thatwere altered in the PSTVd

Int

U257A-infected tomato plants.

Cell Expansion Was Inhibited in Stems and Leaves of PSTVd

Int

U257A-Infected Plants

Plant development results from cell division to increase cellnumbers, from growth to increase cell size, and from differenti-

ation to form specialized cells/tissues (Steeves and Sussex,1989). The restricted shoot development in PSTVd

Int

U257A-infected tomato could be attributable to inhibited cell divisionleading to the production of fewer cells, inhibited growth, ab-normal differentiation of tissues, or a combination of any ofthese processes. Distinguishing between these possibilitiesshould help us understand the cellular processes that were af-fected specifically to cause restricted shoot development andthe appearance of the flat-top symptom.

Figure 5. Accumulation Levels of PSTVd Variants in Infected Tomato Plants and Protoplasts.

RNA gel blot analysis, using total RNA from mock-inoculated and PSTVd-infected tomato plants and protoplasts, was performed with 32P-labeled anti-sense PSTVd probes under high-stringency conditions (see Methods for details). The 5.8S rRNA visualized by ethidium bromide staining indicates equalloading of RNA for all lanes. A 2-ng marker (414 nucleotides) was loaded in lane M. In (A) and (C), C and L denote circular and linear PSTVd, respectively.(A) RNA gel blot analysis of accumulation levels of PSTVd variants in infected tomato plants. Each RNA sample was prepared from a pool of equiva-lent systemic leaves collected from six infected plants at 1-week intervals for up to 6 weeks.(B) Quantitative comparisons of the accumulation levels of plus-strand circular RNAs of PSTVd variants in tomato plants during the 6-week infectionperiod. Each time point represents mean values � SE calculated from three biologically duplicate experiments.(C) Accumulation levels of PSTVd variants in tomato protoplasts. Protoplasts were prepared from tomato leaves and inoculated with PSTVd tran-scripts by electroporation. Protoplasts were harvested for RNA extraction at 3 days after inoculation.

1366 The Plant Cell

We examined the number, size, and type of cells in the stemsof mock-inoculated and PSTVdInt- and PSTVdIntU257A-infectedtomato plants at 3 weeks after inoculation. To ensure that theanalyses were conducted on organs at comparable develop-mental stages in tomato plants, we focused on the fifth inter-nodes from the bottom (i.e., the first internode being the oneimmediately above the cotyledons). This internode was chosenbecause it was the one closest to the shoot apex that still couldbe identified reliably in the PSTVdIntU257A-infected plants. Asshown in Figure 8, the length of the fifth internodes fromPSTVdIntU257A-infected tomato plants was approximately one-third to one-half of that from the mock-inoculated and PSTVdInt-infected plants, even though these plants should be at similardevelopmental stages. The fifth leaf also was much smaller inthe PSTVdIntU257A-infected plants.

Cytological analyses revealed that all tissue types, includingepidermis, cortex, vasculature, and pith, were present in thefifth internodes of PSTVdIntU257A-infected plants, as in theother plants (Figure 9). The numbers of cells in a transversesection were similar in PSTVdInt- and PSTVdIntU257A-infectedand mock-inoculated plants (Table 1). However, the cells in to-mato plants infected with PSTVdIntU257A were approximatelyhalf the size of those in the other plants (Table 1, Figure 9). Inthe fifth leaf of PSTVdIntU257A-infected plants, all cell types

were present and the cell size was smaller than that in the otherplants (Figure 10).

The cytological data suggest that a major cause of the re-duced internode length and leaf size was restricted cell growth.Cell division and differentiation were not affected visibly. Togain further evidence at the molecular level for the inhibited cellgrowth, we analyzed the expression levels of LeExp2 in the fifthinternodes and leaves of mock-inoculated, PSTVdInt-infected,and PSTVdIntU257A-infected tomato plants. LeExp2 encodesan expansin that has been suggested to play a critical role incell expansion in rapidly growing organs, based on its expres-sion patterns in tomato (Reinhardt et al., 1998; Caderas et al.,2000; Catalá et al., 2000). We cloned a 248-bp fragment (nucle-otides 855 to 1102) corresponding to the 3� untranslated regionof LeExp2 from tomato using primers synthesized based on thepublished LeExp2 sequences (Caderas et al., 2000). Thecloned LeExp2 fragment was used as a template for the in vitrosynthesis of radiolabeled riboprobes. RNA gel blot analysisshowed that LeExp2 expression was repressed severely inleaves and young stems of tomato infected with PSTVdIntU257Abut less so in tomato infected with PSTVdInt (Figure 11). As aninternal control, the expression of glyceraldehyde-3-phosphatedehydrogenase (GAPDH) was analyzed and found to be unalteredin any organs (Figure 11). Thus, suppressed LeExp2 expression

Figure 6. Localization of PSTVd in Shoot Apices of Infected Tomato Plants by in Situ Hybridization.

(A) and (B) Serial sections showing the absence of PSTVdInt from the SAM. Viroid signal is present in the vascular tissues (arrows) as well as in othercells in the subapical region.(C) and (D) Serial sections showing the absence of PSTVdIntU257A from the SAM. Viroid signal is present in the vascular tissues as well as in othercells in the subapical region.

RNA Inhibition of Cell Growth 1367

Figure 7. Localization of PSTVd in Leaf and Stem Tissues of Infected Tomato Plants by in Situ Hybridization.

(A) and (B) Transverse (A) and paradermal (B) leaf sections showing the presence of PSTVdInt in all tissue types. Cx, cortex; Ep, epidermis; Ms, meso-phyll; Vas, vascular tissue.(C) and (D) Transverse (C) and paradermal (D) leaf sections showing the presence of PSTVdIntU257A in all tissue types.(E) and (F) Transverse (E) and longitudinal (F) stem sections showing the presence of PSTVdInt in all tissue types. Pi, pith.(G) and (H) Transverse (G) and longitudinal (H) stem sections showing the presence of PSTVdIntU257A in all tissue types.Arrows point to PSTVd localization signals.

1368 The Plant Cell

was specific and was correlated positively with restricted cell ex-pansion in the internodes and leaves of PSTVdIntU257A-infectedtomato plants.

DISCUSSION

A Viroid Can Evolve Multiple Pathogenicity Determinants Spanning the RNA Genome

All nucleotide substitutions in PSTVd that influence symptomexpression have been mapped to the classic pathogenicity do-main (Schnölzer et al., 1985; Owens et al., 1991, 1995, 1996;Hammond, 1992; Gruner et al., 1996). The CCR of PSTVd haslong been thought to be involved mainly in replication (Keeseand Symons, 1985) and host range determination (Wasseneggeret al., 1996). Thus, our finding that a single U257A substitutionin the PSTVd CCR confers symptoms such as severe growthstunting, flat top, and premature death is quite unexpected.Considering the dramatic symptoms that the U257A substitu-tion caused in tomato, it is surprising that this nucleotide sub-

stitution was not reported in previous studies. One explanationmight be that this mutation has adverse effects on viroid repli-cation, so that viroid progeny carrying this mutation disappearrapidly from a host cell. However, our studies showed that thismutation does not affect viroid accumulation in tomato and canbe maintained stably in the infected plants.

Importantly, the U257A substitution did not alter PSTVd ac-cumulation levels in single cells or in a whole plant. This findingis consistent with previous work showing no close correlationbetween accumulation levels and the severity of pathogenicityof some other PSTVd isolates (Schnölzer et al., 1985; Owens etal., 1991; Hammond, 1992). Our results further showed that theU257A substitution does not alter cellular localization patternsof PSTVd. Based on these findings, we conclude that theU257A substitution alters molecular interactions between theviroid and host factors to cause the flat-top symptom. Becausethe U257A substitution does not appear to alter the PSTVdstructure, we postulate that A257 in the CCR functions inde-pendently of the classic pathogenicity domain as a separatepathogenicity determinant.

In testing the role of individual viroid structural domains inpathogenicity, Sano et al. (1992) constructed interspecific chi-meras by exchanging domains between Tomato apical stunt vi-roid and Citrus exocortis viroid, both in the family Pospivi-roidae. Infection studies showed that except for the CCR, allother viroid domains contribute to symptom expression. Here,we demonstrated that the nucleotide sequence of PSTVd CCRalso plays a role in pathogenicity. Together, these data suggestthat there is no definitive viroid pathogenicity domain. Sponta-neous mutations in a viroid genome create diverse sequenceand structural variants. As long as a sequence or structuralchange does not compromise the ability of the viroid variant toreplicate and move systemically, its interaction with a host fac-tor(s) to perturb cellular functions can lead to symptom expres-sion. In this regard, a viroid can serve as a powerful source ma-terial for the study of how an RNA can evolve particularsequences/structures through nucleotide changes to achievedistinct biological functions.

A Viroid Motif Can Have Multiple Functions

This study, like previous studies, reveals the remarkable capac-ity of a viroid RNA motif to possess multiple and diverse func-tions. Subtle nucleotide changes can occur to introduce newfunctions without altering RNA structure or compromising otherfunctions. Loop E of PSTVd CCR is involved directly in pro-cessing of longer-than-unit-length linear plus strands of PSTVdinto unit-length circular molecules (Baumstark et al., 1997). Italso is involved in host adaptation (Wassenegger et al., 1996).We showed that the role of loop E in host adaptation could beattributed at least partially to its regulation of transcription lev-els (Qi and Ding, 2002). Specifically, the U257A or C259U sub-stitution in this motif, although not altering the PSTVd structureor processing capacity, enhances transcription levels �5- to10-fold in tobacco cells (Qi and Ding, 2002). The present studyshows that the U257A substitution can specifically cause theflat-top and growth-stunting symptoms in tomato. Thus, loop Eof PSTVd functions in processing, transcription, and pathoge-

Figure 8. Size Comparison of Fifth Internodes and Fifth Leaves ofMock-Inoculated and PSTVd-Infected Tomato Plants.

(A) The fifth internode from PSTVdIntU257A-infected tomato is one-thirdto one-half the size of those from mock-inoculated and PSTVdInt-infected plants.(B) The fifth leaf of PSTVdIntU257A-infected plant is much smaller thanthose of mock-inoculated and PSTVdInt-infected plants.

RNA Inhibition of Cell Growth 1369

nicity. PSTVd loop E’s role in pathogenicity is similar to the role ofa tetraloop of Chrysanthemum chlorotic mottle viroid (CChMVd) inpathogenicity. A UUUC82-85→GAAA substitution in the tetra-loop converted the symptomatic strain CChMVd-S to a non-symptomatic strain, CChMVd-NS (de la Peña et al., 1999).

Loop E is present in a wide range of RNAs (Branch et al.,1985; Wimberly et al., 1993; Szewczak and Moore, 1995; Leontisand Westhof, 1998a) and functions as an important motif inRNA–RNA and RNA–protein interactions (Correll et al., 1997;Leontis and Westhof, 1998a, 1998b; Gongadze et al., 1999;Hampel and Burke, 2001). This loop may contain submotifs(Leontis and Westhof, 1998c). Based on these observations,two alternative mechanisms can be speculated to account forthe pleiotropic functions of PSTVd loop E. First, submotifs ofloop E mediate interactions with different host factors to ac-complish separate functions. Second, different metastable struc-

tures are formed via interactions of loop E submotifs with otherviroid regions, and these metastable structures then interactwith different host factors to perform distinct functions. Eluci-dating how loop E performs multiple functions should providevaluable insights about RNA structure-function relationships.

Roles of RNA Sequences in Viroid and Viral Pathogenicity

Viroid variants with slight nucleotide sequence differences of-ten cause different degrees of symptom severity (Schnölzeret al., 1985; Owens, 1990; Owens et al., 1991, 1995, 1996;Hammond, 1992; kori et al., 2001). Such differences may re-sult in differences in viroid structure. For instance, nucleotidedifferences in the virulence-modulating region within the patho-genicity domain of PSTVdMild, PSTVdInt, and PSTVdRG1 lead todifferent degrees of bending of the virulence-modulating region

S c

Figure 9. Structure of the Fifth Internodes of Mock-Inoculated and PSTVd-Infected Tomato Plants.

(A) and (B) Transverse (A) and longitudinal (B) sections of the fifth internodes from mock-inoculated plants. Cx, cortex; Ep, epidermis; Pi, pith; Vas,vascular tissue.(C) and (D) Transverse (C) and longitudinal (D) sections of the fifth internodes from PSTVdInt-infected plants.(E) and (F) Transverse (E) and longitudinal (F) sections of the fifth internodes from PSTVdIntU257A-infected plants. All tissue types are present. Thecell size is one-third to one-half of that from the other plants.

1370 The Plant Cell

(Owens et al., 1996; Schmitz and Riesner, 1998). Such differentbending may influence binding with host factors for symptomexpression (Owens et al., 1996; Schmitz and Riesner, 1998).

The nucleotide at position 257 must be A, and not U, G, or C,in the PSTVdInt background to confer the flat-top symptom.This represents an example of how subtle nucleotide sequencedifferences in the viroid genome can confer distinct interactionswith host factors. Thus, there are numerous possibilities for aviroid RNA to generate sequence variants, which may or maynot assume higher level structures, that can interact with andperturb host functions and lead to disease formation. There isalready an example of how subtle nucleotide differences with-out obvious effect on the PSTVd secondary structure have sig-nificant impact on interactions with host factors. Diener et al.(1993) showed that PSTVdMild and PSTVdSevere, which differ bytwo nucleotides in the lower portion of the classic pathogenic-ity domain, have a 10-fold difference in activating the inter-feron-induced, double-stranded RNA-activated protein kinase(P68). Whether such differential activation of P68 is directly re-sponsible for the development of some viroid symptoms re-mains to be investigated.

Direct RNA–host factor interactions for symptom expressionmay not be limited to viroid infection. Although most studies ofviral pathogenesis have revealed roles of specific viral proteins(reviewed by Hull, 2002; Maule et al., 2002), a few studiesdemonstrate that noncoding elements of a viral genomic RNA(Rodríguez-Cerezo et al., 1991; Fernandez et al., 1999) or a viralsatellite RNA (Taliansky and Robinson, 1997; Taliansky et al.,1998) can control symptom development in an infected host, po-tentially via direct interactions between the RNA sequences/struc-tures and host factors. Thus, direct RNA–host factor interactionsmay be a fundamental mechanism of pathogenicity shared by vi-roids, viruses, and virus-associated RNAs such as satellite RNAs.

The Possibility of Linking Viroid Structure, Host Gene Expression, Cellular Response, and Symptom Expression

Studies of plant viral and viroid diseases have produced muchinformation about the pathogen determinants of symptom ex-

pression (Hull, 2002). The global gene expression pattern in aninfected host has been examined by microarray analysis duringviral infection (Whitham et al., 2003) and by macroarray analysisduring viroid infection (Itaya et al., 2002). In general, there is alack of understanding of how the altered expression of a partic-

Table 1. Cell Number and Size in the Fifth Internodes ofMock-Inoculated and PSTVd-Infected Tomato Plants

InoculumCortexCell No.

PithCell No.a Pith Cell Size (�m � �m)b

Water 1486 � 173 1735 � 185 78.2 � 10.3 � 89.1 � 12.6PSTVdInt 1432 � 145 1676 � 203 60.0 � 5.4 � 75.2 � 6.7PSTVdIntU257A 1551 � 218 1786 � 156 33.5 � 5.6 � 55.8 � 8.1

aThe number of cells represents the number on a transverse section.The number in each case is the mean � SE calculated from cell countsobtained from four complete transverse sections of the internode.bThe size was measured as length (along the length of the internode) andwidth (across the width of the internode) of a cell in a longitudinal sectionin each case. For each longitudinal section, the cell length was calcu-lated by dividing the entire length of the section by the total number ofcells along the length. Similarly, the cell width was calculated by dividingthe entire width of the section by the total number of cells along thewidth. Four sections were used to measure the cell size in each case.

Figure 10. Structure of the Fifth Leaves of Mock-Inoculated andPSTVd-Infected Tomato Plants.

(A) Transverse section of a leaf from a mock-inoculated plant.(B) Transverse section of a leaf from a PSTVdInt-infected plant.(C) Transverse section of a leaf from a PSTVdIntU257A-infected plant.All tissue types, including cortex (Cx), epidermis (Ep), mesophyll (Ms),and vascular tissue (Vas), are present in all leaves. The size of cells fromthe PSTVdIntU257A-infected plant is one-third to one-half of that fromthe other plants.

RNA Inhibition of Cell Growth 1371

ular host gene causes a specific symptom. In particular, the al-tered cellular processes that serve as a link between alteredhost gene expression and a specific symptom type are poorlystudied or understood.

In this study, we have attempted to learn about the cellularprocesses that link a particular viroid pathogenicity determinantwith host symptom expression, using a combination of molecu-lar, cellular, biophysical, and whole-plant approaches. Our re-sults demonstrate that PSTVdIntU257A infection causes specifi-cally restricted cell expansion but not obviously cell divisionand differentiation. Consequently, this leads to internode short-ening and leaf miniaturization. Downregulated expression ofLeExp2 provides a molecular marker for inhibited cell expan-sion. This finding is fully consistent with the observations thatLeExp2 is expressed in rapidly growing parts of hypocotyls,stems, and leaves and with the hypothesis that the product ofthis gene functions to loosen cell walls to permit cell expansion(Reinhardt et al., 1998; Caderas et al., 2000; Catalá et al.,2000). We recognize that LeExp2 likely functions in concertwith other proteins to regulate cell wall expansion (Caderas etal., 2000; Catalá et al., 2000) and that LeExp2 is likely one ofthe many genes whose expression is altered in PSTVdIntU257A-infected plants. Furthermore, altered expression of LeExp2 mayoccur downstream of the cascade of altered gene expressionthat leads to the flat-top symptom. Nonetheless, our data pro-vide a concrete example of how a pathogen-elicited symptomat the whole-plant level can be linked to the disturbance of aspecific cellular process, which in turn is correlated positivelywith the repressed expression of a cellular gene implicated ex-plicitly in this process. Further studies now are possible to in-vestigate the upstream genes that regulate LeExp2 expressionand the primary targets of PSTVdIntU257A.

Viroid Infection: A Useful Model in Which to Study Direct RNA Regulation of Cellular Processes

Without encoding proteins, a viroid RNA interacts directly withcellular factors to alter cellular functions that lead to changes ingrowth and development. Thus, a viroid is a de facto noncodingand regulatory RNA. Cellular noncoding and regulatory RNAs oc-cur widely in organisms ranging from animals (Lagos-Quintanaet al., 2001; Lau et al., 2001; Lee and Ambros, 2001) to plants(Llave et al., 2002a; Reinhart et al., 2002). Recent studies haveunderscored the potential importance of such RNAs in the reg-ulation of plant growth and development (Llave et al., 2002b;Rhoades et al., 2002).

Disruption of microRNA-controlled developmental processesis at least partly responsible for the formation of viral symptoms(Kasschau et al., 2003). Thus, viroid pathogenicity in all casesand viral pathogenicity in at least some cases share a commonground with the regulation of gene expression by cellular non-coding and regulatory RNAs: the direct effect of RNA on cellu-lar functions. Studying the mechanisms of viroid and viralpathogenicity should provide fundamental insights into themechanisms of RNA-regulated gene expression. Well-under-stood viroid structure/function relationships also may permitthe engineering of RNA motifs with specific functions as re-search tools to study various biological processes.

METHODS

Plant Material and Growth Conditions

Tomato plants (Lycopersicon esculentum cv Rutgers) were grown in agrowth chamber with 27/22�C day/night temperatures and a 14-h/10-hlight/dark cycle.

PSTVd Variants and cDNA Cloning

Plasmids pRZ6-2, pRZ:Mild, and pRZ:RG1, which have cDNAs ofPSTVdInt (Gross et al., 1978), PSTVdMild (Schnölzer et al., 1985), andPSTVdRG1 (Zimmat et al., 1990), respectively, flanked by double ri-bozymes, were constructed by Hu et al. (1997) and kindly provided to usby Robert Owens (U.S. Department of Agriculture/Agricultural ResearchService, Beltsville, MD). Construction of plasmids pRZ:IntC259U andpRZ:IntU257A was described by Zhu et al. (2002). Plasmids pRZ:KF440-2,pRZ:KFU257A, pRZ:IntU257G, pRZ:IntU257C, pRZ:IntC259U/U257A,pInter(�), and pInter(�) were described by Qi and Ding (2002).

pRZ:MildU257A and pRZ:RG1U257A with the U257A substitution andpRZ:IntA135G with the A135G substitution were generated by site-directed mutagenesis using the Quickchange Site-Directed MutagenesisKit (Stratagene, La Jolla, CA) according to the manufacturer’s instruc-tions. The reactions were performed with plasmids pRZ:Mild, pRZ:RG1,or pRZ:6-2 and complementary primers with sequences correspondingor complementary to PSTVd sequences with the desired mutation. Theintroduced mutations were verified by sequencing.

cDNA Cloning of Tomato LeExp2 and Glyceraldehyde-3-Phosphate Dehydrogenase Gene Fragments

A 248-bp fragment corresponding to the 3� untranslated region ofLeExp2 (corresponding to nucleotides 855 to 1102) (Caderas et al.,

Figure 11. Expression Levels of the LeExp2 Gene in the Fifth Leavesand Fifth Internodes of Mock-Inoculated and PSTVd-Infected TomatoPlants.

RNA gel blot analysis, using total RNA from fifth leaves and internodesof mock-inoculated, PSTVdInt-infected, and PSTVdIntU257A-infectedplants, was performed with 32P-labeled LeExp2-specific probes underhigh-stringency conditions (top gels). The membrane was stripped andreprobed with 32P-labeled GAPDH-specific probes (middle gels). The28S rRNA visualized by ethidium bromide staining indicates approxi-mately equal loading of RNA for all lanes (bottom gels).

1372 The Plant Cell

2000) was amplified by reverse transcriptase–mediated PCR and thencloned into pGEM-T vector (Promega, Madison, WI) in the antisense ori-entation. This gave rise to plasmid pLeExp2(�), which was used as thetemplate to produce antisense LeExp2 riboprobes. Plasmid pGAPDH(�)was provided by Asuka Itaya (Department of Plant Biology and PlantBiotechnology Center, Ohio State University, Columbus) and used toproduce antisense glyceraldehyde-3-phosphate dehydrogenase (GAPDH)riboprobes. pGAPDH(�) has a 500-bp fragment of tomato GAPDH (cor-responding to nucleotides 276 to 775), which was inserted into pCR2.1-TOPO vector (Invitrogen, Carlsbad, CA).

In Vitro Transcription

To prepare in vitro transcripts of the PSTVd variants for inoculation, theplasmids described above containing PSTVd cDNAs flanked by doubleribozymes were linearized with HindIII and used as templates for in vitrotranscription using the T7 MEGAscript Kit (Ambion, Austin, TX) accord-ing to the manufacturer’s directions. To obtain 32P-labeled PSTVd tran-scripts for temperature gradient gel electrophoresis (TGGE) analysis,UTP was replaced by -32P-UTP in in vitro transcription reactions. SpeI-linearized pInter(�) was used as a template to synthesize marker RNAsof 414 nucleotides that contain monomeric, linear (�)-PSTVd sequencesand 55 nucleotides from the vector.

To prepare riboprobes for RNA gel blot analysis or in situ hybridiza-tion, -32P– or digoxigenin-UTP–labeled antisense riboprobes wereprepared by in vitro transcription using the T7 Maxiscript kit (Ambion) ac-cording to the methods recommended by the manufacturer using SpeI-linearized pInter(�), pLeExp2(�), and pGAPDH(�) as templates.

After transcription, DNA templates were removed by digestion withRNase-free DNase I, and RNA transcripts were purified using the MEGA-clear kit (Ambion). Nonradioactive RNA transcripts were quantified byUV spectrometry.

Plant and Protoplast Inoculation

Water and in vitro transcripts of PSTVd variants were used as inocula.Ten microliters of water or PSTVd transcripts (100 ng/�L) was rubbedonto each carborundum-dusted cotyledon of each 6-day-old tomatoseedling with a Pasteur pipette. Six plants were inoculated with water ortranscripts of each PSTVd variant in one set of experiments. At least fourindependent sets of experiments were conducted to ensure the repro-ducibility of symptom expression caused by each PSTVd variant.

Protoplasts were prepared from young leaves of 3-week-old tomatoplants as described by Mühlbach and Sänger (1977) with the followingmodifications: solution I [0.5 M mannitol and 3.6 mM 2-(N-morpholino)-ethanesulfonic acid, pH 5.9] was used instead of 0.78 M mannitol and0.5% potassium dextran sulfate; protoplasts were resuspended in solu-tion II (solution I plus 0.1 mM CaCl2) to a density of 2 � 106 protoplasts/mL and kept on ice for 1 h before inoculation. The electroporationmethod used to inoculate protoplasts with PSTVd transcripts was es-sentially as described by Qi and Ding (2002). After inoculation, proto-plasts were cultured in solution III (Schenk and Hildebrandt medium[Sigma, St. Louis, MO] plus 0.45 M mannitol). At 3 days after inoculation,the protoplasts were harvested for RNA extraction.

RNA Extraction and RNA Gel Blot Analysis

Total RNA was isolated from plants or protoplasts using the RNeasyPlant Mini Kit (Qiagen, Valencia, CA) and quantified by UV spectrometry.For detection of PSTVd accumulation, RNA aliquots (10 �g each) werefractionated by electrophoresis at 55�C on 5% polyacrylamide gels con-taining 1� TBE buffer (90 mM Tris-borate, 2 mM EDTA, pH 8.0) and 8 Murea. To determine the expression levels of LeExp2 and GAPDH, RNA ali-

quots were subjected to 3-(N-morpholino)-propanesulfonic acid–form-aldehyde agarose gel electrophoresis (Sambrook et al., 1989). Afterelectrophoresis, the gels were stained with ethidium bromide and exam-ined under UV light to determine the integrity and loading of the samequantity of RNA samples. The RNAs were transferred to HybondXL ny-lon membranes (Amersham Biosciences, Piscataway, NJ) using a vac-uum-blotting system (Amersham). Hybridizations were performed at65�C with ULTRAhyb reagent (Ambion) and in vitro transcribed -32P-UTP–labeled antisense riboprobes. The membranes were washed at68�C and exposed to a Storage Phosphor Screen (Kodak, Rochester,NY). Quantification of radioactivity was performed with Molecular ImagerFX using Quantity One-4.1.1 software (Bio-Rad, Hercules, CA).

Sequencing of RNA Progeny

The protocols for preparing cDNAs of the PSTVd progeny isolated frominfected plants or protoplasts were essentially as described by Qi andDing (2002). cDNAs of PSTVd RNA were amplified by reverse tran-scriptase–mediated PCR and sequenced in both directions using theABI377 DNA sequencer (Perkin-Elmer, Boston, MA) at the DNA Se-quencing Facility at Ohio State University.

Tissue Processing

Tissue processing was performed as described previously (Zhu et al.,2001). Tomato samples were fixed in 10% formaldehyde, 50% ethanol,and 5% acetic acid, dehydrated, and embedded in paraffin (Electron Mi-croscopy Sciences, Fort Washington, PA). Sections (10 �m) were ob-tained with a rotary microtome (model HM 340 E; Microm International,Walldorf, Germany).

Histology

Tissue sections were dewaxed in xylene, treated sequentially with anhy-drous isopropanol and 70% isopropanol, stained with 0.5% (w/v) safraninin 95% isopropanol, and rinsed with 50% isopropanol. Stained sectionswere mounted and examined with a Nikon Eclipse 600 light microscope(Nikon, Tokyo, Japan). Images were captured with a SPOT 2 Slidercharge-coupled device camera and the associated software (DiagnosticsInstruments, Sterling Heights, MI). Image processing and figure prepara-tion were performed using Adobe Photoshop (Mountain View, CA).

In Situ Hybridization

In situ hybridization was performed as described previously (Zhu et al.,2001) using digoxigenin-labeled antisense PSTVd riboprobes. After hy-bridization, the sections were examined and photographed with theNikon Eclipse 600 light microscope as described above.

Analysis of PSTVd RNAs by Perpendicular TGGE

Preparation of PSTVd RNAs for TGGE was performed essentially as de-scribed by Owens et al. (1996). 32P-labeled linear unit-length PSTVdRNAs were purified by electrophoresis at 55�C on 5% polyacrylamidegels containing 1� TBE buffer and 8 M urea and circularized by incuba-tion with wheat germ extract (Promega). Perpendicular TGGE analysis of32P-labeled circular and linear PSTVd RNAs was performed using aTGGE system (Biometra, Göttingen, Germany) according to the instruc-tions of the manufacturer using 5% polyacrylamide gels (acrylamide:bisacrylamide [30:1], 0.2� TBE, and 5 mM NaCl) and running buffer (0.2�

TBE and 5 mM NaCl). The gel electrophoresis procedure was as follows:(1) prerunning at 25�C and 200 V for 5 min; (2) loading the mixture of

RNA Inhibition of Cell Growth 1373

PSTVd RNAs (15,000 cpm each); (3) running at 25�C and 250 V for 6 min;(4) establishing a 25 to 60�C temperature gradient in 3 min; and (5) con-tinued running at 250 V for 25 min. After electrophoresis, the gel wasfixed in 10% ethanol and 0.5% acetic acid, dried, and subjected to au-toradiography.

Upon request, all novel materials described in this article will be madeavailable in a timely manner for noncommercial research purposes.

Accession Number

The GenBank accession number for tomato GAPDH is U97257.

ACKNOWLEDGMENTS

Yali Zhu is acknowledged for the initial observation of PSTVdIntU257A-induced tomato symptoms and technical assistance. We express ourgratitude to Yan Xun for technical assistance. We thank Asuka Itaya forproviding plasmid pGAPDH(�) and for providing many insightful sug-gestions and comments. We are especially indebted to Labrepco forproviding us with the TGGE unit and to Robert Owens for advice onTGGE experiments. We thank David Bisaro, Robert Owens, Richard Nelson,Jack Morris, Jeanmarie Verchot-Lubicz, and Xianfeng Xu for helpful dis-cussions. This work was supported by grants from the U.S. Departmentof Agriculture National Research Initiative Competitive Grants Program(Grants 97-35303-4519 and 2001-35303-11073).

Received February 26, 2003; accepted April 2, 2003.

REFERENCES

Baumstark, T., Schröder, A.R.W., and Riesner, D. (1997). Viroid pro-cessing: Switch from cleavage to ligation is driven by a change from atetraloop to a loop E conformation. EMBO J. 16, 599–610.

Branch, A.D., Benenfeld, B.J., and Robertson, H.D. (1985). Ultravioletlight-induced crosslinking reveals a unique region of local tertiarystructure in potato spindle tuber viroid and HeLa 5S RNA. Proc. Natl.Acad. Sci. USA 82, 6590–6594.

Caderas, D., Muster, M., Vogler, H., Mandel, T., Rose, J.K.C.,McQueen-Mason, S., and Kuhlemeier, C. (2000). Limited correla-tion between expansion gene expression and elongation growth rate.Plant Physiol. 123, 1399–1413.

Catalá, C., Rose, J.K.C., and Bennett, A.B. (2000). Auxin-regulatedgenes encoding cell wall modifying proteins are expressed duringearly tomato fruit growth. Plant Physiol. 122, 527–534.

Correll, C.C., Freeborn, B., Moore, P.B., and Steitz, T.A. (1997). Met-als, motifs, and recognition in the crystal structure of a 5S rRNAdomain. Cell 91, 705–712.

de la Peña, M., Navarro, B., and Flores, R. (1999). Mapping the molec-ular determinant of pathogenicity in a hammerhead viroid: A tetraloopwithin the in vivo branched RNA conformation. Proc. Natl. Acad. Sci.USA 96, 9960–9965.

Diener, T.O., Hammond, R.W., Black, T., and Katze, M.G. (1993).Mechanisms of viroid pathogenesis: Differential activation of theinterferon-induced, double-stranded RNA-activated, Mr 68 000 pro-tein kinase by viroid strains of varying pathogenicity. Biochemie 75,533–538.

Fernandez, I., Candresse, T., Le Gall, O., and Dunez, J. (1999). The 5�

noncoding region of grapevine chrome mosaic nepovirus RNA-2 trig-gers a necrotic response on three Nicotiana spp. Mol. Plant-MicrobeInteract. 12, 337–344.

Flores, R., Di Serio, F., and Hernández, C. (1997). Viroids: The non-encoding genomes. Semin. Virol. 8, 65–73.

Flores, R., Randles, J.W., Bar-Josef, M., and Diener, T.O. (2000).Subviral agents: Viroids. In Virus Taxonomy, Seventh Report of the Inter-national Committee on Taxonomy of Viruses, M.H.V. van Regenmortel,C.M. Fauquet, D.H.L. Bishop, E.B. Carstens, M.K. Estes, S.M. Lemon, J.Manilof, M.A. Mayo, D.J. McGeoch, C.R. Pringle, and R.B. Wickner, eds(San Diego, CA: Academic Press), pp. 1009–1024.

Gast, F.-U., Kempe, D., Ludwig, R., Spieker, R.L., and Sänger, H.L.(1996). Secondary structure probing of potato spindle tuber viroid(PSTVd) and sequence comparison with other small pathogenic RNAreplicons provides evidence for central non-canonical base-pairs,large A-rich loops, and a terminal branch. J. Mol. Biol. 262, 652–670.

Gongadze, G.M., Meshcheryakov, V.A., Serganov, A.A., Fomenkova,N.P., Mudrik, E.S., Jonsson, B.-H., Liljas, A., Nikonov, S.V., andGarber, M.B. (1999). N-terminal domain, residues 1-91, of ribosomalprotein TL5 from Thermus thermophilus binds specifically andstrongly to the regions of 5S rRNA containing loop E. FEBS Lett. 451,51–55.

Gross, H.J., Domdey, H., Lossow, C., Jank, P., Raba, M., Alberty, H.,and Sänger, H.L. (1978). Nucleotide sequence and secondary struc-ture of potato spindle tuber viroid. Nature 273, 203–208.

Gruner, E., Fels, A., Qu, F., Zimmat, R., Steger, G., and Riesner, D.(1996). Interdependence of pathogenicity and replicability with potatospindle tuber viroid. Virology 209, 60–69.

Hammond, R.W. (1992). Analysis of the virulence modulating region ofpotato spindle tuber viroid (PSTVd) by site-directed mutagenesis.Virology 187, 654–662.

Hampel, K.J., and Burke, J.M. (2001). A conformational change in the“loop E-like” motif of the hairpin ribozyme is coincidental with domaindocking and is essential for catalysis. Biochemistry 40, 3723–3729.

Hu, Y., Feldstein, P.A., Hammond, J., Hammond, R.W., Bottino, P.J.,and Owens, R.A. (1997). Destabilization of potato spindle tuber viroidby mutations in the left terminal loop. J. Gen. Virol. 78, 1199–1206.

Hull, R. (2002). Matthew’s Plant Virology, 4th ed. (San Diego, CA: Aca-demic Press).

Itaya, A., Matsuda, Y., Gonzales, R.A., Nelson, R.S., and Ding, B.(2002). Potato spindle tuber viroid strains of different pathogenicityinduces and suppresses expression of common and unique genes ininfected tomato. Mol. Plant-Microbe Interact. 15, 990–999.

Kasschau, K., Xie, Z., Allen, E., Llave, C., Chapman, E.J., Krizan,K.A., and Carrington, J.C. (2003). P1/HC-Pro, a viral suppressor ofRNA silencing, interferes with Arabidopsis development and miRNAfunction. Dev. Cell 4, 205–217.

Keese, P., and Symons, R.H. (1985). Domains in viroids: Evidence ofintermolecular RNA rearrangements and their contribution to viroidevolution. Proc. Natl. Acad. Sci. USA 82, 4582–4586.

Lagos-Quintana, M., Rauhut, R., Lendeckel, W., and Tuschl, T.(2001). Identification of novel genes coding for small expressedRNAs. Science 294, 853–858.

Lau, N.C., Lim, L.P., Weinstein, E.G., and Bartel, D.P. (2001). Anabundant class of tiny RNAs with probable regulatory roles in Cae-norhabditis elegans. Science 294, 858–862.

Lee, R.C., and Ambros, V. (2001). An extensive class of small RNAs inCaenorhabditis elegans. Science 294, 862–864.

Leontis, N.B., and Westhof, E. (1998a). A common motif organizes thestructure of multi-helix loops in 16 S and 23 S ribosomal RNAs. J.Mol. Biol. 283, 571–583.

Leontis, N.B., and Westhof, E. (1998b). Conserved geometric base-pairing patterns in RNA. Q. Rev. Biophys. 31, 399–455.

Leontis, N.B., and Westhof, E. (1998c). The 5S rRNA loop E: Chemicalprobing and phylogenetic data versus crystal structure. RNA 4, 1134–1153.

1374 The Plant Cell

Llave, C., Kasschau, K.D., Rector, M.A., and Carrington, J.C.(2002a). Endogenous and silencing-associated small RNAs in plants.Plant Cell 14, 1605–1619.

Llave, C., Xie, Z., Kasschau, K.D., and Carrington, J.C. (2002b).Cleavage of Scarecrow-like mRNA targets directed by a class of Ara-bidopsis miRNA. Science 297, 2053–2056.

Loss, P., Schmitz, M., Steger, G., and Riesner, D. (1991). Formationof a thermodynamically metastable structure containing hairpin II iscritical for infectivity of potato spindle tuber viroid RNA. EMBO J. 10,719–727.

Maule, A., Leh, V., and Lederer, C. (2002). The dialogue betweenviruses and hosts in compatible interactions. Curr. Opin. Plant Biol. 5,279–284.

Mühlbach, H.-P., and Sänger, H.L. (1977). Multiplication of cucumberpale fruit viroid in inoculated tomato leaf protoplasts. J. Gen. Virol. 35,377–386.

Owens, R.A. (1990). Mutational analysis of viroid pathogenicity: Tomatoapical stunt viroid. Mol. Plant-Microbe Interact. 3, 374–380.

Owens, R.A., Chen, W., Hu, Y., and Hsu, Y.-H. (1995). Suppression ofpotato spindle tuber viroid replication and symptom expression bymutations which stabilize the pathogenicity domain. Virology 208,554–564.

Owens, R.A., Steger, G., Hu, Y., Fels, A., Hammond, R.W., andRiesner, D. (1996). RNA structural features responsible for potatospindle tuber viroid pathogenicity. Virology 222, 144–158.

Owens, R.A., Thompson, S.M., and Steger, G. (1991). Effects of ran-dom mutagenesis upon potato spindle tuber viroid replication andsymptom expression. Virology 185, 18–31.

Qi, Y., and Ding, B. (2002). Replication of Potato spindle tuber viroid incultured cells of tobacco and Nicotiana benthamiana: The role of spe-cific nucleotides in determining replication levels for host adaptation.Virology 302, 445–456.

Qu, F., Heinrich, C., Loss, P., Steger, G., Tien, P., and Riesner, D.(1993). Multiple pathways of reversion in viroids for conservation ofstructural elements. EMBO J. 12, 2129–2139.

Reinhardt, D., Wittwer, F., Mandel, T., and Kuhlemeier, C. (1998).Localized upregulation of a new expansin gene predicts the site ofleaf formation in the tomato meristem. Plant Cell 10, 1427–1437.

Reinhart, B.J., Weinstein, E.G., Rhoades, M.W., Bartel, B., and Bartel,D.P. (2002). MicroRNAs in plants. Genes Dev. 16, 1616–1626.

Rhoades, M.W., Reinhart, B.J., Lim, L.P., Burge, C.B., Bartel, B., andBartel, D.P. (2002). Prediction of plant microRNA targets. Cell 110,513–520.

Riesner, D. (1991). Viroids: From thermodynamics to cellular structureand function. Mol. Plant-Microbe Interact. 4, 122–131.

Rodríguez-Cerezo, E., Klein, P.G., and Shaw, J.G. (1991). A determi-nant of disease symptom severity is located in the 3�-terminal non-coding region of the RNA of a plant virus. Proc. Natl. Acad. Sci. USA88, 9863–9867.

Rosenbaum, V., and Riesner, D. (1987). Temperature gradient gelelectrophoresis: Thermodynamic analysis of nucleic acids and proteinin purified form and in cellular extracts. Biophys. Chem. 26, 235–246.

Sambrook, J., Fritsch, E.F., and Maniatis, T. (1989). Molecular Clon-ing: A Laboratory Manual. (Cold Spring Harbor, NY: Cold Spring Har-bor Laboratory Press).

Sano, T., Candresse, T., Hammond, R.W., Diener, T.O., and Owens,R.A. (1992). Identification of multiple structural domains regulatingviroid pathogenicity. Proc. Natl. Acad. Sci. USA 89, 10104–10108.

Schmitz, A., and Riesner, D. (1998). Correlation between bending ofthe VM region and pathogenicity of different potato spindle tuberviroid strains. RNA 4, 1295–1303.

Schnölzer, M., Haas, B., Ramm, K., Hofmann, H., and Sänger, H.L.(1985). Correlation between structure and pathogenicity of potatospindle tuber viroid (PSTVd). EMBO J. 4, 2181–2190.

Schröder, A.R., and Riesner, D. (2002). Detection and analysis of hair-pin II, an essential metastable structural element in viroid replicationintermediates. Nucleic Acids Res. 30, 3349–3359.

kori , D., Conerly, M., Szychowski, J.A., and Semancik, J.S. (2001).CEVd-induced symptom modification as a response to a host-spe-cific temperature-sensitive reaction. Virology 280, 115–123.

Steeves, T.A., and Sussex, I.M. (1989). Patterns in Plant Development,2nd ed. (Oxford, UK: Cambridge University Press).

Storz, G. (2002). An expanding universe of noncoding RNAs. Science296, 1260–1263.

Sussex, I.M. (1989). Developmental programming of the shoot mer-istem. Cell 56, 225–229.

Szewczak, A.A., and Moore, P.B. (1995). The sarcin/ricin loop, a mod-ular RNA. J. Mol. Biol. 247, 81–98.

Taliansky, M.E., and Robinson, D.J. (1997). Trans-acting untranslatedelements of groundnut rosette virus satellite RNA are involved insymptom production. J. Gen. Virol. 78, 1277–1285.

Taliansky, M.E., Ryabov, E.V., Robinson, D.J., and Palukaitis, P.(1998). Tomato cell death mediated by complementary plant viralsatellite RNA sequences. Mol. Plant-Microbe Interact. 11, 1214–1222.

Wassenegger, M., Spieker, R.L., Thalmeir, S., Gast, F.-U., Riedel, L.,and Sänger, H.L. (1996). A single nucleotide substitution convertspotato spindle tuber viroid (PSTVd) from a noninfectious to an infec-tious RNA for Nicotiana tabacum. Virology 226, 191–197.

Water, A.E., Turner, D.H., Kim, J., Lyttle, M.H., Muller, P., Mathews,D.H., and Zuker, M. (1994). Coaxial stacking of helixes enhancesbinding of oligoribonucleotides and improves predictions of RNAfolding. Proc. Natl. Acad. Sci. USA 91, 9218–9222.

Whitham, S.A., Quan, S., Chang, H.-S., Cooper, B., Estes, B., Zhu, T.,Wang, X., and Hou, Y.-M. (2003). Diverse RNA viruses elicit theexpression of common sets of genes in susceptible Arabidopsisthaliana plants. Plant J. 33, 271–283.

Wimberly, B., Varani, G., and Tinoco, I., Jr. (1993). The conformation ofloop E of eukaryotic 5S ribosomal RNA. Biochemistry 32, 1078–1087.

Zhu, Y., Green, L., Woo, Y.-M., Owens, R.A., and Ding, B. (2001). Cel-lular basis of potato spindle tuber viroid systemic movement. Virology279, 69–77.

Zhu, Y., Qi, Y., Xun, Y., Owens, R., and Ding, B. (2002). Movement ofPotato spindle tuber viroid reveals regulatory points of phloem-medi-ated RNA traffic. Plant Physiol. 130, 138–146.

Zimmat, R., Gruner, R., Hecker, R., Steger, G., and Riesner, D.(1990). Analysis of mutations in viroid RNA by non-denaturing andtemperature gradient gel electrophoresis. In Proceedings of the 6thConversation in Biomolecular Stereodynamics, Vol. 3, R. Sarma andM. Sarma, eds (Schenectady, NY: Adenine Press), pp. 339–357.

S c

DOI 10.1105/tpc.011585; originally published online May 1, 2003; 2003;15;1360-1374Plant Cell

Yijun Qi and Biao DingNoncoding Viroid RNA

Inhibition of Cell Growth and Shoot Development by a Specific Nucleotide Sequence in a

 This information is current as of September 3, 2018

 

References /content/15/6/1360.full.html#ref-list-1

This article cites 58 articles, 20 of which can be accessed free at:

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists