RAPID COMMUNICATION

Activation of hypersensitive response genes in the absenceof pathogens in transgenic tobacco plants expressinga rice small GTPase

Received: 7 May 2003 / Accepted: 25 June 2003 / Published online: 23 August 2003� Springer-Verlag 2003

Abstract Transgenic tobacco (Nicotiana tabacum L.)plants constitutively expressing a rice (Oryza sativa L.)gene encoding a small GTPase, rgp1, showed markedresistance to tobacco mosaic virus (TMV) infectioncompared with the wild type [H. Sano et al. (1994)Proc Natl Acad Sci USA 91:10556–10560]. In order toexamine the gene expression profile, the temperature-shift method was adopted to hyper-activate the N-geneinducing the hypersensitive response (HR), and tran-scripts of 11 representative HR genes were analyzed.In transgenic and wild-type plants, transcripts of 10genes were induced during the HR; however, in mostcases, their expression level was higher in the formerthan in the latter. Mock treatment of transgenic plantsalso efficiently induced transcripts of 8 out of 11 genesafter temperature shift, indicating that their activationis mediated by the N-gene. Salicylic acid and itsglucoside-conjugates were induced in both transgenicand wild-type plants, but their quantity in the formerwas unusually higher than in the latter. These resultssuggest that expression of rgp1 positively influencedthe signaling pathway of the HR, resulting in higherinduction of salicylates. This possibly caused a‘‘priming effect’’ that hyper-activates the HR genesthrough the N-gene without TMV infection. It wasthus conceivable that, despite a structural similarity tothe Rab-family of GTPases, which function in mem-brane trafficking, rgp1 might participate in the signaltransduction pathway of the HR.

Keywords Hypersensitive response Æ Rab-family smallGTPase Æ Tobacco mosaic virus Æ Nicotiana

Abbreviations HR: hypersensitive response Æ SA: sali-cylic acid Æ SAG: salicylic acid b-glucoside Æ TMV:tobacco mosaic virus

Introduction

Small GTPases, often called Ras-subfamily G-proteins,are considered to be molecular switches for many bio-chemical reactions (Hall and Zerial 1995). To date,hundreds of such proteins have been identified in variousorganisms, and classified into several major subgroups(Hall and Zerial 1995). The Ras-subfamily proteins areinvolved in signal transduction for cell growth and dif-ferentiation. Rho-subfamily proteins are thought tofunction in cytoskeleton formation, and the Rabsubfamily in membrane trafficking. The Arf-, Sar- andRan-family members are also considered to be involvedin intracellular transport. Among the subfamilies, Rabconstitutes the largest group, which is ubiquitously dis-tributed among organisms, including plants and yeast.Curiously, Ras-subfamily proteins have not been foundin plants. As they regulate expression of many genes inanimals by mediating phosphorylation cascades, theirlack suggests the presence of so far unidentified alter-native signaling systems in plants. The major function ofRab-subfamily proteins is the regulation of vesiculartransport, involved, for example, in exocytosis andendocytosis. In Arabidopsis, 57 genes encoding RabGTPases have been identified and classified into 8groups (Rutherford and Moore 2002), suggesting thattheir roles are diverse and complex (Vernoud et al.2003). It remains to be clarified whether they are in-volved in signaling pathways for stress responses but,judging from available data, such a function appearsunlikely.

We have previously identified from rice plants a smallGTPase, rgp1, whose amino acid sequence shows theclosest similarity to rab11/ypt3-group proteins (Sanoand Youseffian 1991). Transgenic tobacco plants

Planta (2003) 217: 993–997DOI 10.1007/s00425-003-1092-6

Hiroshi Yoda Æ Hiroshi Sano

H. Yoda Æ H. Sano (&)Research and Education Center for Genetic Information,Nara Institute of Science and Technology,630-0192 Nara, JapanE-mail: sano@gtc.aist-nara.ac.jpFax: +81-743-725659

expressing rgp1 showed a dwarf phenotype with elevatedcytokinin levels, and also resistance to tobacco mosaicvirus (TMV) infection (Sano et al. 1994). Further anal-yses indicated that they responded unusually tomechanical wounding by producing large amounts ofsalicylic acid (Sano et al. 1994). These observationssuggested that, in rgp1 transgenic tobacco, there is cross-talk between the signaling pathways for wounding andpathogen responses. It was thus speculated that rgp1might function as a regulatory factor in stress responses,probably modulating membrane trafficking (Sano andOhashi 1995). However, the exact cause–effect relation-ship between its expression and increased resistance toTMV was not clear. In the present study, we furtheranalyzed the pathogen response of the rgp1 transgenicplants, and obtained results suggesting that rgp1 maypositively function in the defense signal transductionpathway during TMV infection.

Materials and methods

Plant material and TMV treatment

Transgenic tobacco (Nicotiana tabacum L. cv. Xanthi nc) express-ing rice (Oryza sativa L.) rgp1 was constructed as previously de-scribed (Kamada et al. 1992), and T3 progeny of the original A2strain were used in the present study. As controls, wild-type to-bacco plants (N. tabacum cv. Xanthi nc) were employed. Bothcontrol and transgenic plants were grown in a cabinet at 23 �Cunder a 14/10 h light/dark cycle. About 8-week-old mature leaveswere detached, inoculated with or without (mock treatment) TMV(10 lg/ml) in phosphate buffer (pH 7.2) using Carborundum(Mesh 600) as described by Sano et al. (1994). Leaves were incu-bated under continuous light at 30 �C for 48 h, and then at 20 �C,allowing the plants to super-express the hypersensitive response(HR). Leaves were frozen in liquid nitrogen immediately afterappropriate time intervals, and stored at )80 �C until extraction ofRNA.

RNA isolation and gel-blot analysis

Total cellular RNAs were isolated by the acid guanidine–phenol–chloroform method (Chomczynski and Sacchi 1987). Twenty-lgaliquots were fractionated by formaldehyde/agarose gel electro-phoresis and transferred to nylon membranes (Hybond-N;Amersham). After baking for 2 h at 80 �C, the membranes weresubjected to hybridization with appropriate 32P-labeled probesunder medium-stringency conditions at 42 �C for 16 h in a solutioncontaining 1 mM EDTA, 0.5% SDS, 50 mM Tris–HCl (pH 7.5),1 · Denhardt’s, 3 · SSC, 50% formamide and 10% dextran sulfate.After successive washing, membranes were exposed to either BASor X-ray film (Kodak).

Isolation and estimation of salicylic acid and salicylic acidb-glucoside

Salicylic acid (SA) and salicylic acid b-glucoside (SAG) wereextracted and quantified essentially as described by Raskin et al.(1989). Briefly, 1 g of frozen tissue was subjected to methanolextraction, and resulting extracts were divided into two fractionsfor estimation of SA and SAG. SAG was enzymatically hydrolyzedto SA and glucose with b-glucosidase (2.2 units/ml). SA wasmeasured using an HPLC apparatus equipped with a Scanning

Fluorescence Detector (Waters 474) with a TSKgel ODS-80Tscartridge column (4.6 mm · 15.0 cm; Tohso, Tokyo) and a TSK-guardgel ODS-80Ts guard column (3.2 mm · 1.5 cm; Tohso) witha flow rate of 0.5 ml/min. SA concentration was determined withan excitation wavelength of 313 nm and an emission wavelength of405 nm. A standard curve for SA was determined by measuringauthentic samples (1, 10, 50, 100 ng).

Results and discussion

Activation of HR-related genes in the absenceof a pathogen

Our previous studies (Sano et al. 1994, 1996) revealedthat transgenic tobacco expressing rgp1 unusuallyresponds to mechanical wounding by producing SA andaccumulating PR gene transcripts, and also displayingincreased resistance to TMV infection. In order toidentify elements of the enhanced resistance, we furtherexamined genes whose transcripts might be differentiallyinduced between the wild type and T3 progeny of theoriginal A2 transgenic plants. After TMV inoculation, asample leaf was subjected to the temperature-shift tohyper-activate the N-gene, so that the host recognizesTMV infection and better induces the HR (Yoda et al.2002). Total RNA was then isolated at various timepoints, and RNA gel-blot hybridization was performedwith cDNAs of genes whose transcripts were known tobe induced during the HR in wild-type plants (Yodaet al. 2002). The results showed that among 11 genestested, 10 were up-regulated and one down-regulatedduring the HR in both A2 and control plants, althoughthe extent of their induction differed (Fig. 1). Forexample, the levels of induced transcripts for PAL,TIZZ and ACCS were markedly higher in A2 than incontrol plants (Fig. 1). An unexpected finding, however,was that 8 out of 11 genes also responded to the mocktreatment. This was typically seen for PI-II, PR-1a andPAL, whose transcripts began to accumulate as early as6 h after the temperature shift (Fig. 1). Notably, HRR1transcripts were down-regulated by mock treatment,showing a similar decrease as in the case of TMVinfection (Fig. 1). These results indicated two distinctaspects of A2 plants. First, HR-related genes can beactivated in the absence of TMV infection, and second,such activation is still under the influence of the N-gene.It was thus likely that the high resistance of A2 plants toTMV might be due to the accelerated expression of HRgenes, which were additively activated by TMV andenvironmental cues.

Accumulation of SA during the HR

Salicylic acid is considered to be a signaling molecule,which possibly triggers expression of some genesinvolved in the HR pathway. Since we previously foundSA to be induced inA2 plants uponwounding (Sano et al.1994), we speculated that the induction of HR-related

994

gene transcripts by the mock treatment might possibly bedue to the modulation of signal transduction pathway byelevated levels of SA following initial Carborundumwounding. Subsequent assays showed that, under mockconditions, SA content was constantly low and equal inboth A2 and wild-type plants, whereas SAG was almost2-fold higher in the former than in the latter (Fig. 2).Such an elevated level of SAG did not appear be due tothe temperature shift but to be inherent in A2 plants.Upon HR induction by TMV infection, SA was equally

induced in both A2 and wild-type plants, but its level wasnearly 2-fold higher in A2 than in the control (Fig. 2,upper panel). This pattern was distinct for SAG induc-tion, showing more than 2-fold higher amounts in A2than in the control, although the induction pattern itselfwas again similar between the two (Fig. 2, lower panel).Thus, it became clear that induction of SA and SAGduring the HR is qualitatively similar but quantitativelydifferent in A2 and wild-type plants.

Priming in the A2 plants

The elevated level of SA in A2 plants may induce a stateof alert that has increasingly been appreciated in recenttimes, and is called ‘‘priming’’ (Zimmerli et al. 2000;Conrath et al. 2002). Priming refers to the fact that de-fense induction by plant tissues is not necessarily a directresponse to the pathogens. Rather, upon subsequentinfection of the tissue, the defense reactions are theninduced more rapidly and/or to a greater extent.Intriguingly, primed plant tissue also has increasedresponsiveness to wounding, as has been demonstratedfor Arabidopsis (Kohler et al. 2002) and tobacco (Muret al. 1996).

When shifted to normal temperature after infection,A2 plants produce more SA and SAG than does the wild

Fig. 1 Time course analysis of transcript accumulation. Healthyleaves of wild-type (WT) or rgp1-transgenic (A2) tobacco (Nicoti-ana tabacum) plants were harvested, treated with buffer alone(Mock) or with buffer containing TMV (TMV) and maintained inan incubator at 30 �C for 48 h, and then at 20 �C for appropriatetimes post temperature shift (h). Total RNAs were isolated andRNA gel-blot analysis was conducted with 32P-labeled probes forgenes encoding proteinase inhibitor II (PI-II), pathogenesis relatedprotein-1a (PR-1a) phenylalanine ammonia-lyase (PAL), WRKYtranscription factor (TIZZ; Yoda et al. 2002), aminocyclopropanecarboxylate synthase (ACCS), basic pathogenesis related protein-1(Basic PR-1) and ornithine decarboxylase (ODC). Unidentifiedgenes up-regulated during the HR are HRR2 (unknown protein),HRR6 (unknown protein) and HRR7 (HSR203 J), and HRR1(possibly encoding subunit III of photosystem I; Yoda et al. 2002)was down-regulated. As an internal standard for RNA loading, thetranscript level of the actin gene was estimated

Fig. 2 Time course analysis of SA and SAG accumulation duringthe HR. Healthy leaves of wild-type or rgp1-transgenic tobaccoplants were treated with or without TMV as described in the legendto Fig. 1, and contents of SA (upper panel ) and SAG (lower panel )were measured at the indicated time points after temperature shift.Examined samples were mock-treated wild-type (open circles),mock-treated A2 (closed circles), TMV-treated wild-type (opentriangles) and TMV-treated A2 (closed triangles)

995

type, suggesting that the plants have become highlyprimed. Subsequently, the temperature shift mayhyper-activate the N gene and/or its downstreamsignaling pathways and super-induce both HR genesand the HR itself, which is associated with tissuedamage. Due to the latter, some other genes also becomebetter activated in the plant. In the mock-treated A2plants, wound stimulation with Carborundum results inproduction of SA, which may induce priming. This mayexplain the unusual expression of HR genes withoutTMV infection in A2 plants.

In either case, unusual accumulation of SA and SAGin A2 plants appears to be associated with priming.However, the mechanism maintaining the high back-ground level of SAG in A2 plants is not clear, but oneexplanation is that SA, which is produced by woundingof A2 plants (Sano et al. 1994), may be converted intoSAG. If this were the case, the SA effects are even self-amplifying. SA alone is not able to induce most HRgenes, such as PAL (Kohler et al. 2002; Mur et al. 1996).However, due to SA-mediated priming there isaugmented activation of subsequently wound- or TMV-induced genes (Fig. 1). Thus, it is tempting to speculatethat the initial priming by unusually enhanced SAproduction upon wounding caused the increasedproduction of SA, which might be responsible for theobserved potentiation of HR genes and the improvedinduction of the HR. This situation is consistent with theaugmented HR and HR-associated gene activation inSA-treated soybean cells (Shirasu et al. 1997).

Function of the small GTPase during the HR

Amino acid sequence analysis indicated that rgp1 is mostclosely related to yeast Ypt3, mammalian Rab11 andArabidopsis RabA2, which are known to play importantroles in vesicle transport from the Golgi apparatus to theendosomal/prevacuolar compartment (Rutherford andMoore 2002). Physiological analyses with transgenicplants suggested that Rab11 homologs might be involvedin pathogen defense, cell elongation, fruit softening andmorphogenesis (Ueda and Nakano 2002). It is generallyconsidered that such effects are possibly due to alterationsin the endocytotic pathway in cells, although an influenceon gene expression cannot be excluded. For example, thePra3 protein from pea is reported to act to integrate lightexposure with the brassinosteroid signaling pathway inthe etiolation response (Kang et al. 2001). However, it hasbeen suggested that this relationship might be indirectand due to a defect in vacuole function (Schumacher et al.1999; Inaba et al. 2002). Therefore, it remains contro-versial whether or not Rab-subfamily proteins, includingRab11 homologs, are involved in signal transductionpathways for environmental response.

In this context, the present study suggests a novel fea-ture of small GTPase proteins, showing that rgp1 possiblyfunctions in the HR signaling pathway under influence ofthe N-gene. Two explanations are conceivable for this

specific action: first, disturbance of membrane traffickingindirectly causes an alteration of the HR; second, rgp1 isdirectly involved in the HR signaling system. The firstassumption is consistent with the previous finding thattransgenic plants expressing plant Rab genes show moreor less similar phenotypes, such as dwarfism, blanchingand abnormal tissue structures (Kamada et al. 1992;Asupuria et al. 1995). This implies that the Rab proteinshave pleiotropic effects on plant differentiation anddevelopment. Perhaps the stress response pathway is alsoaffected by endocytotic trafficking and alterations in thelatter may result in an altered HR. However, the questionarises of how rgp1might confer such a specific influence ata specific point among numerous signaling pathways. Inthis sense, the second assumption is more attractive toexplain our findings, and we speculate that rgp1 directlyparticipates inHRsignaling in a similarmanner to theRasproteins. Although supporting observations for this ideahave yet to be reported, it is conceivable that, in plants,Rab or Rab-like proteins substitute for Ras of otherorganisms as signaling components. Indeed, the smallGTPases of many plants have been classified into the Rabsubfamily simply based on similarities in amino acidsequence, but few have been examined for physiologicalroles.More physiological analyses using transgenic plantsshould help to clarify this point.

Acknowledgements The authors are grateful to Drs. N. Koizumiand Y. Yamaguchi (Nara Institute of Science and Technology) forvaluable suggestions and discussion, and Dr. Malcolm A. Moore(Intermal, Nagoya) for critical reading of the manuscript. Thiswork was supported by a grant from the Research for the FutureProgram (JSPS-RFTF00L01604) of the Japan Society for thePromotion of Science.

References

Aspuria ET, Anai T, Fujii N, Ueda T, Miyoshi M, Matsui M,Uchimiya H (1995) Phenotypic instability of transgenic tobaccoplants and their progenies expressing Arabidopsis thaliana smallGTP-binding protein genes. Mol Gen Genet 246:509–513

Chomczynski P, Sacchi N (1987) Single-step of RNA isolation byacid guanidinium thiocyanate–phenol–chloroform extraction.Anal Biochem 162:156–159

Conrath U, Pieterse CM, Mauch-Mani B (2002) Priming in plant–pathogen interactions. Trends Plant Sci 7:210–216

Hall A, Zerial M (1995) Overviews of the Ras superfamily of smallGTPases. In: Zerial M, Huber LA (eds) Guidebook to the smallGTPases. Oxford University Press, Oxford, pp 3–11

Inaba T, Nagano Y, Nagasaki T, Sasaki Y (2002) Distinct locali-zation of two closely related Ypt3/Rab11 proteins on the traf-ficking pathway in higher plants. J Biol Chem 277:9183–9188

Kamada I, Yamauchi S, Youssefian S, Sano H (1992). Transgenictobacco plants expressing rgp1, a gene encoding a ras-relatedGTP-binding protein from rice, show distinct morphologicalcharacteristics. Plant J 2:799–807

Kang J-G, Yun J, Kim DH, Chung KS, Fujioka S, Kim JI, DaeHW, Yoshida S, Takatsuto S, Song PS, Park CM (2001) Lightand brassinosteroid signals are integrated via a dark-inducedsmall G protein in etiolated seedling growth. Cell 105:625–636

Kohler A, Schwindling S, Conrath U (2002) Benzothiadiazole-in-duced priming for potentiated responses to pathogen infection,wounding, and infiltration of water into leaves requires theNPR1/NIM1 gene in Arabidopsis. Plant Physiol 128:1046–1056

996

Mur LA, Naylor G, Warner SA, Sugars JM, White RF, Draper J(1996) Salicylic acid potentiates defense gene expression in tis-sue exhibiting acquired resistance to pathogen attack. Plant J9:559–571

Raskin I, Turner IM, Melander WR (1989) Regulation of heatproduction in the inflorescences of an Arum lily by endogenoussalicylic acid. Proc Natl Acad Sci USA 86:2214–2218

Rutherford S, Moore I (2002) The Arabidopsis Rab GTPase family:another enigma variation. Curr Opin Plant Biol 5:518–528

Sano H, Ohashi Y (1995) Involvement of small GTP-bindingproteins in defense signal-transduction pathways of higherplants. Proc Natl Acad Sci USA 92:4138–4144

Sano H, Youssefian S (1991) A novel ras-related rgp1 geneencoding a GTP-binding protein has reduced expression in 5-azacytidine-induced dwarf rice. Mol Gen Genet 228:227–232

Sano H, Seo S, Orudgev E, Youssefian S. Ishizuka K, Ohashi Y(1994) Expression of a small GTP-binding protein gene in to-bacco plants elevates cytokinin levels and abnormally inducessalicylic acid in response to wounding and increases resistanceto tobacco mosaic virus infection. Proc Natl Acad Sci USA91:10556–10560

Sano H. Seo S, Koizumi N, Niki T, Iwamura H, Ohashi Y (1996)Regulation by cytokinins of endogenous levels of jasmonic andsalicylic acids in mechanically wounded tobacco plants. PlantCell Physiol 37:762–769

Schumacher K, Vafeados D, McCarthy M, Sze H, Wilkins T,Chory J (1999) The Arabidopsis det3 mutant reveals a centralrole for the vacuolar H+-ATPase in plant growth and devel-opment. Genes Dev 13:3259–3270

Shirasu K, Nakajima H, Rajasekhar VK, Dixon R, Lamb C (1997)Salicylic acid potentiates an agonist-dependent gain controlthat pathogen signals in the activation of defense mechanisms.Plant Cell 9:261–270

Ueda T, Nakano A (2002) Vesicular traffic: an integral part ofplant life. Curr Opin Plant Biol 5:513–517

Vernoud V, Horton AC, Yang Z, Nielsen E (2003) Analysis of thesmall GTPase gene superfamily of Arabidopsis. Plant Physiol131:1191–1208

Yoda H, Ogawa M, Yamaguchi Y, Koizumi N, Sano H (2002)Identification of early responsive genes associated with thehypersensitive response and properties of a WRKY-type tran-scription factor in tobacco plants upon tobacco mosaic virusinfection. Mol Genet Genom 267:154–161

Zimmerli L, Jakab G, Metraux JP, Mauch-Main B (2000) Potenti-ation of pathogen-specific defense mechanisms in Arabidopsis byb-aminobutyric acid. Proc Natl Acad Sci USA 97:12920–12925

997