Overexpression of the Disease Resistance Gene Pto in Tomato Induces Gene … · Overexpression of the Disease Resistance Gene Pto in Tomato Induces Gene Expression Changes Similar

Overexpression of the Disease Resistance Gene Pto inTomato Induces Gene Expression Changes Similar toImmune Responses in Human and Fruitfly1[w]

Kirankumar S. Mysore*, Mark D. D’Ascenzo, Xiaohua He2, and Gregory B. Martin

Plant Biology Division, Samuel Roberts Noble Foundation, Ardmore, Oklahoma 73402 (K.S.M.); BoyceThompson Institute for Plant Research, Ithaca, New York 14853 (K.S.M., M.D.D., X.H., G.B.M.); andDepartment of Plant Pathology, Cornell University, Ithaca, New York 14853 (G.B.M.)

The Pto gene encodes a serine/threonine protein kinase that confers resistance in tomato (Lycopersicon esculentum) toPseudomonas syringae pv tomato strains that express the type III effector protein AvrPto. Constitutive overexpression of Ptoin tomato, in the absence of AvrPto, activates defense responses and confers resistance to several diverse bacterial and fungalplant pathogens. We have used a series of gene discovery and expression profiling methods to examine the effect of Ptooverexpression in tomato leaves. Analysis of the tomato expressed sequence tag database and suppression subtractivehybridization identified 600 genes that were potentially differentially expressed in Pto-overexpressing tomato plantscompared with a sibling line lacking Pto. By using cDNA microarrays, we verified changes in expression of many of thesegenes at various time points after inoculation with P. syringae pv tomato (avrPto) of the resistant Pto-overexpressing line andthe susceptible sibling line. The combination of these three approaches led to the identification of 223 POR (Pto overex-pression responsive) genes. Strikingly, 40% of the genes induced in the Pto-overexpressing plants previously have beenshown to be differentially expressed during the human (Homo sapiens) and/or fruitfly (Drosophila melanogaster) immuneresponses.

Both plants and animals are continually exposed topathogens, and, as a result, they have evolved intri-cate defense mechanisms to recognize and defendthemselves against a wide array of these disease-causing agents. Recent studies have revealed com-mon mechanisms of pathogen virulence and hostresistance underlying plant and animal diseases(Cohn et al., 2001; Staskawicz et al., 2001; Nurnbergerand Brunner, 2002). For example, bacterial pathogensof both plants and animals deliver virulence proteinsinto the host cytoplasm via the type III secretionsystem (He, 1998; Galan and Collmer, 1999).Pathogen-associated molecular patterns that are sim-ilar to those activating immune responses in animalshave been shown to mediate activation of plant de-fenses (Nurnberger and Brunner, 2002). Some plantdisease resistance (R) proteins share motifs with sig-naling components of immune response pathways in

mammals and invertebrates (Staskawicz et al., 2001).A striking similarity underlying the molecular mech-anisms of pathogen recognition in plants and animalsis suggested by the recent isolation of the intracellu-lar receptors NOD1 and NOD2 involved in human(Homo sapiens) Crohn’s disease (Ogura et al., 2001a).In common with a large family of plant R proteins,the NOD1/2 proteins contain a putative nucleotide-binding site and a region of Leu-rich repeats (Oguraet al., 2001b).

In tomato (Lycopersicon esculentum), the R gene Ptoencodes a Ser/Thr kinase and confers resistanceagainst strains of Pseudomonas syringae pv tomato thatexpress the effector proteins AvrPto or AvrPtoB(Martin et al., 1993; Kim et al., 2002). Overexpressionof Pto in tomato under control of the cauliflowermosaic virus (CaMV) 35S promoter has been shownto activate defense responses in the absence of patho-gen inoculation. These responses include PR geneexpression, an increased level of salicylic acid (SA),and development of spontaneous microscopic lesionssimilar to the hypersensitive response (HR; Tang etal., 1999; Xiao et al., 2001). Pto-overexpressing plantsshow resistance not only to P. syringae pv tomato butalso to Xanthomonas campestris pv vesicatoria and tothe fungal pathogen Cladosporium fulvum. Thus, theyexhibit an exaggerated form of the typical plant de-fense response, making them a useful model to dis-sect the complicated plant responses that occur dur-ing race-nonspecific resistance to several pathogens(Tang et al., 1999). Recently, Xiao et al. (2001) iden-

1 This work was partly supported by Boyce Thompson Institute(innovation grant to K.S.M.), by the Noble Foundation (grants toK.S.M.), and by the National Science Foundation (Plant Genomicsgrant nos. IBN–9872617 and IBN– 0109633 to G.B.M.).

2 Present Address: U.S. Department of Agriculture, AgriculturalResearch Service, Western Regional Research Center, Albany, CA94710.

[w] The online version of this article contains Web-only data. Thesupplemental material is available at http://www.plantphysiol.org.

* Corresponding author; e-mail [email protected]; fax 580–224–6692.

Article, publication date, and citation information can be foundat www.plantphysiol.org/cgi/doi/10.1104/pp.103.022731.

Plant Physiology, August 2003, Vol. 132, pp. 1901–1912, www.plantphysiol.org © 2003 American Society of Plant Biologists 1901 www.plantphysiol.orgon April 23, 2020 - Published by Downloaded from Copyright © 2003 American Society of Plant Biologists. All rights reserved.

http://www.plantphysiol.org

tified a limited number of genes that are expressedconstitutively (without inoculation by P. syringae pvtomato expressing avrPto) in the Pto-overexpressingtomato line. Many of the genes encode known PRproteins, confirming that overexpression of Pto acti-vates bona fide defense responses in plants.

We have now used suppression subtractive hybrid-ization (SSH; Diatchenko et al., 1996), an in-depthanalysis of the tomato expressed sequence tag (EST)database, and cDNA microarray analysis to charac-terize more fully the alterations in gene expressioncaused by the overexpression of Pto. Interestingly,over 40% of the genes that are highly expressed inPto-overexpressing plants either with or without P.syringae pv tomato infection are also known to beinduced in human and/or fruitfly (Drosophila mela-nogaster) during various defense responses. Thesefindings indicate that many downstream defense re-sponses in plants, humans, and invertebrates, as seenfor certain recognition and signaling components, areshared and, thus, are either highly conserved or mayhave arisen by convergent evolution.

RESULTS

Use of in Silico EST Subtraction to IdentifyGenes That Are Differentially Expressed uponP. syringae pv tomato Inoculation in TomatoLeaves Overexpressing Pto

As one method of identifying genes that might bedifferentially expressed in response to inoculationwith P. syringae pv tomato(avrPto), we analyzed thetomato EST database for sequences derived from lineR11-12 that expresses the Pto gene from the CaMV35S promoter and a sibling line R11-13 that has seg-regated away this transgene. As described in “Mate-rials and Methods,” both lines were inoculated withthe avirulent P. syringae pv tomato(avrPto), and leaveswere harvested at various time points afterward. Weanalyzed 10,872 EST sequences from the R11-12 (R)and R11-13 (S) libraries (5,316 for R library and 5,556for S library). A total of 6,921 of these EST sequencesassembled into 1,414 contigs that contained eitherESTs all derived from one of the libraries (116 contigswith at least three ESTs; 42 contained ESTs only fromthe R library, whereas 74 contigs contained ESTs onlyfrom the S library; Fig. 1) or a mix of ESTs from bothlibraries (1,298 contigs). The 3,953 singleton ESTswere excluded from further analysis. Of the 1,298contigs with mixed R and S sequences, we identified132 cases where the EST counts within a contig dif-fered by more than 3-fold between R and S libraries(Fig. 1). An additional 51 contigs were identified thatencoded known defense genes and that had countsbetween the libraries differing by more than 2-fold.To test if the genes corresponding to these ESTs weredifferentially expressed in P. syringae pv tomato-inoculated R11-12 and R11-13 leaves, we selected arepresentative cDNA clone from each of the 299 con-

tigs (116 cases with at least three ESTs coming onlyfrom either R or S library, 132 cases with �3-folddifferences, and the 51 cases with �2-fold differ-ences) for microarray analysis (see below).

Use of cDNA Subtraction to Identify GenesThat Are Differentially Expressed uponP. syringae pv tomato(avrPto) Inoculation inTomato Leaves Overexpressing Pto

In a second approach to identify differentially ex-pressed genes, we performed cDNA subtraction, anopen-architecture gene expression profiling tech-nique. By using SSH (Diatchenko et al., 1996), we“forward” subtracted cDNAs derived from R11-12plants from cDNAs derived from R11-13 plants andvice versa (“reverse” subtraction) after inoculationwith P. syringae pv tomato(avrPto). In the forwardsubtraction experiments, we used RNA isolated fromleaves collected at 4 and 8 h after inoculation toidentify genes expressed or overexpressed in R11-12(tester) as compared with R11-13 (driver). Similarly,we used reverse subtraction to identify genes whoseexpression was suppressed in R11-12 (driver) as com-pared with R11-13 (tester). We followed up the sub-traction procedure in two ways. First, we directlycloned the resulting PCR products and sequencedabout 65 clones from each of the forward and reversesubtractions. Second, we radioactively labeled thesubtracted PCR products and used them as a pooled

Figure 1. Identification of differentially expressed genes by in silicoEST subtraction. ESTs derived from the R11-12 (R) and the R11-13 (S)libraries were assembled into contigs. Within each contig, the num-ber of ESTs derived from the R or S libraries was then calculated.Those contigs (with at least three ESTs) with ESTs from only the Rlibrary are shown on the left, and those with ESTs from only the Slibrary are shown on the right. Contigs that contained mixtures ofESTs from both libraries are shown in the middle. Fifty-four of thesecontigs contained 3-fold or more ESTs from the R than S library,whereas 78 contigs contained 3-fold or more ESTs from the S than theR library. The remainder of the contigs (1,298) contained mixtures ofESTs from both libraries but in less than 3-fold ratios.

Mysore et al.

1902 Plant Physiol. Vol. 132, 2003 www.plantphysiol.orgon April 23, 2020 - Published by Downloaded from Copyright © 2003 American Society of Plant Biologists. All rights reserved.


probe on colony blots containing the DNA from anonredundant EST set developed from the EST se-quences of R and S libraries. Based on the relativehybridization intensities, approximately 250 of thespotted cDNA clones were observed to correspond toabundant sequences in the subtraction libraries. Se-quences from both of these methods were searchedagainst National Center for Biotechnology Information(NCBI) databases (see below). We further examined thedifferential expression of genes identified by SSH byusing cDNA microarrays and RNA blots (see supple-mental data Fig. S1 at http://www.plantphysiol.org).

cDNA Microarray Analysis of Genes Identified by inSilico EST Subtraction and SSH

A total of 600 cDNA clones (some redundancy wasobserved) obtained from SSH and in silico EST sub-traction were selected for microarray analysis. cDNAclones corresponding to known plant defense relatedgenes were also included. Both R11-12 and R11-13plants were inoculated with P. syringae pv toma-to(avrPto), and RNA was isolated at different timepoints (0, 1, 2, 4, and 8 h) afterward. These RNAsamples were used for microarray hybridization.From 299 genes identified by in silico EST subtrac-tion, 159 (53%) of them were differentially expressed(�2-fold) in at least one of the five time points ana-lyzed. Of the 301 clones that were identified by SSH,158 clones (52%) were shown to be differentiallyexpressed (�2-fold) in at least one of the five timepoints. Overall, by combining the approaches of insilico EST subtraction, SSH, and cDNA microarrays,we identified 223 nonredundant genes that were sig-nificantly differentially expressed at one or moretime points between the R11-12 and R11-13 plants(see supplemental data Table S1 at http://www.plantphysiol.org). We designate these genes as POR(Pto overexpression responsive).

Scatter plots were generated to display the differ-ential expression of the POR genes at different timepoints after inoculation (Fig. 2). Interestingly, 106

genes were differentially expressed more than 2-foldat the 0-h time point (without any P. syringae pvtomato inoculation), and the transcripts of most ofthese genes were more abundant in Pto-over-expressing line R11-12 than in the susceptible lineR11-13. Remarkably, the transcript abundance ofsome of the genes (osmotin like, class I chitinase, andbasic 30-kD endochitinase precursor) at 0-h timepoint was greater than 100-fold in R11-12 as com-pared with R11-13. The number of genes with lowertranscript abundance in R11-12 than in R11-13 in-creased markedly at 8 h after inoculation when com-pared with the 0-h time point (Fig. 2). The number ofgenes that were differentially expressed by morethan 10-fold in R11-12 compared with R11-13 wasreduced at 8 h after inoculation when compared withthe 0-h time point. This is not unexpected becausemany defense-related genes are known to be ex-pressed early in incompatible interactions and tothen be very highly expressed later in the compatibleinteraction. Expression data of the 223 POR genes(those genes with �2-fold expression differences)were used for cluster analysis as described below.

Coordinate Regulation of Genes That AreConstitutively Differentially Expressed in LineR11-12 and That Are Differentially Expressed byP. syringae pv tomato(avrPto) Inoculation

The ratios of spot intensities of the 223 POR genesobtained at different time points after inoculation ofR11-12 and R11-13 with P. syringae pv tomato(avrPto)were used to do a hierarchical cluster analysis (Eisenet al., 1998; Fig. 3). From this analysis, we focused onseven clusters (Fig. 3, A–G; see supplemental dataFig. S2 at http://www.plantphysiol.org). Cluster Acontains a variety of genes whose expression is in-duced in R11-12 leaves after inoculation with P. sy-ringae pv tomato(avrPto). Maximum induction ofthese genes was observed at 8 h after inoculation.Some of the genes that belong to this cluster encodeproteins involved in phenylpropanoid metabolism,

Figure 2. Scatter plots showing distribution of POR gene expression ratios at different time points after inoculation. EachcDNA insert was replicated three times on the microarray slides (see “Materials and Methods”). The dots (�2-foldinduction/repression) and white circles (�2-fold induction/repression) correspond to the signal intensity ratios (R11-12:R11-13) of the individual cDNA spots. The x axis represents the spot intensity of Cy3 (R11-12 RNA; experiment), and the y axisrepresents the spot intensity of Cy5 (R11-13 RNA; control). Diagonal blue and brown lines represent �2-fold and �10-foldinduction:repression ratio cutoffs, respectively, relative to the best fit line through the normalized data (middle black line).Number of cDNA clones that are �2-fold or �10-fold induced (red)/repressed (green) is shown.

Overexpression of Pto in Tomato

Plant Physiol. Vol. 132, 2003 1903 www.plantphysiol.orgon April 23, 2020 - Published by Downloaded from Copyright © 2003 American Society of Plant Biologists. All rights reserved.


water transport, cell protection, or stress responses(Fig. S2a). Cluster B genes were highly expressedwithout P. syringae pv tomato(avrPto) inoculation inR11-12 plants when compared with R11-13 plants,and their transcript levels remained almost constantat time points after inoculation. Cluster B genes en-code a variety of proteins, including a glutathioneS-transferase (GST), a ferritin, an AP2 domain tran-scription factor, and some PR genes (Fig. S2a). Thiscluster also contains Pto whose high expression isdriven by the CaMV 35S promoter. Cluster C in-cludes mainly PR genes. It also includes cell pro-tectant genes, an ethylene response factor (ERF) tran-scription factor Pti4, and a WRKY transcription factor(Fig. S2b). These genes were highly expressed inR11-12 plants before pathogen inoculation, and theratio of expression between R11-12 and R11-13 plantsgradually decreased beginning 2 h after inoculation.This expression pattern is due to the induction ofthese genes in R11-13 plants after inoculation. Cluster

D includes a variety of genes including PR genes,HR-associated genes, general stress-related genes,and another ERF transcription factor Pti5 (Fig. 3).These genes were highly expressed in R11-12 plantsbefore pathogen inoculation, and the ratio of expres-sion between R11-12 and R11-13 decreased beginning1 h after inoculation. Again, this is due to the induc-tion of these genes in R11-13 plants at this time point.

Cluster E includes a set of genes whose expressionin R11-12 is suppressed during the resistance re-sponse, in comparison with R11-13 plants, with thesuppression starting at 4 h after inoculation andreaching a maximum at 8 h after inoculation. ClusterE contains mostly genes encoding chloroplast- andphotosynthesis-related proteins (Fig. S2c). Cluster Fcontains genes whose expression was suppressed,without pathogen inoculation, in the R11-12 plantswhen compared with R11-13 plants. Cluster F com-prises a variety of genes that encode proteins liketranslation initiation and elongation factors, ACC ox-

Figure 3. A hierarchical cluster of gene expression of the 223 POR genes over a time course after inoculation by P. syringaepv tomato expressing avrPto. Each gene is represented by a single row of colored bars, and the columns represent the timepoints. Colored bars (red, increased transcript abundance; green, decreased transcript abundance; gray, missing orincomplete data; and black, no difference in transcript abundance) represent the ratio of hybridization measurementsbetween corresponding time points in the R11-12 and R11-13 (treated with avirulent P. syringae pv tomato) samples. Thecluster tree (left) illustrates the nodes of coregulation of gene expression. The clusters A through G are discussed in the text.Average differential expression and the list of genes in clusters D and G are shown. Genes with asterisks are previouslyknown plant defense-related genes that were deliberately included during microarray experiments.

Mysore et al.



idase, prohibitin, proteasome alpha-subunit, andchromatin-associated proteins (Fig. S2c). Expressionof genes in cluster G is very similar to Cluster Eexcept that the suppression of these genes was ob-served only at 8 h after inoculation. Cluster G is alsoenriched with chloroplast- and photosynthesis-related genes (Fig. 3).

Functional Classification of POR Genes

POR genes were classified based on their potentialcellular functions into 21 different groups (Table I).Of the 21 classes, the majority of the genes in 16 of theclasses were “up-regulated” in R11-12 plants.Viewed broadly, they include genes involved in cellprotection (from oxidative stress), cell wall fortifica-tion, hormone responses, HR, general metabolism,known plant defense functions, transport, signaling,ubiquitination, and water transport, and genes en-coding pathogenesis-related proteins, ribosomal pro-teins, stress-related proteins, transcription factors,proteins with no homology to an annotated gene inthe database, and genes encoding proteins of un-known function. Three classes of genes had both up-and down-regulated genes, and these include genesrequired for various other miscellaneous functions,phenylpropanoid metabolism, and senescence-associ-ated proteins. It is interesting to note that almost 19%

of the POR genes were photosynthesis- and chloro-plast-related genes (see supplemental data Table S1 athttp://www.plantphysiol.org). Without any excep-tions, all the photosynthesis- and chloroplast-relatedgenes were “down-regulated” during the resistanceresponse. All photosynthesis- and chloroplast-relatedgenes were suppressed only after pathogen inocula-tion and had maximum suppression at 8 h afterinoculation.

Overexpression of Pto Causes Differential Expression ofGenes That Belong to the Same Functional ClassesIdentified in Tomato Leaves with Pto Expressed from ItsNative Promoter

We compared the POR genes with a recently iden-tified set of genes that is differentially expressed 4 hafter inoculation with avirulent P. syringae pv toma-to(avrPto) in a tomato line expressing Pto from itsnative promoter (RG-PtoR; Mysore et al., 2002). Ap-proximately 22% of POR genes (48 of 223) were rep-resented in the previously identified set of genes(see supplemental data Table S1 at http://www.plantphysiol.org; Mysore et al., 2002). The differen-tial expression of 16 of these 48 genes was muchhigher in Pto-overexpressing plants when comparedwith RG-PtoR plants. Although the other 78% of thegenes were distinct from the previously identified setof genes, they fall into functional classes that aresimilar to the classes of genes identified previously inRG-PtoR plants. Thus, Pto-overexpressing plants ex-press a subset of genes shared with the normal Pto-mediated disease resistance response but also ex-press a distinct set of largely defense-related genes.For example, the majority of genes involved in cellprotection and defense were “up-regulated,” andchloroplast- and photosynthesis-related genes were“down-regulated” during the resistance response inboth RG-PtoR and R11-12 plants after inoculationwith P. syringae pv tomato(avrPto).

Similarities between the Plant Defense Response and theImmune Response of Humans and Fruitfly

Because striking similarities exist among both Rproteins and downstream signaling components ofplants and certain animal and fruitfly proteins withroles in immunity (Cohn et al., 2001; Staskawicz etal., 2001), we decided to compare the set of PORgenes with several expression profiling studies of themammalian/invertebrate immune response (Grego-rio et al., 2001; Huang et al., 2001; Irving et al., 2001).We found that for 62 of 153 POR genes (approximate-ly 40%) that were abundantly expressed in R11-12plants when compared with R11-13 plants, closelyrelated genes (based on putative function and not bysequence similarity) were induced during the humanand/or fruitfly immune responses (Table II). Tran-scripts of most of these genes were abundant in

Table I. Functional classification of POR genes

Functional ClassificationNo. of

Up-regulatedGenesa

No. ofDown-regulated

Genesa

Cell protectant 18 1Cell wall associated 8 –Chloroplast associated – 12Hormone responsive 3 –HR related 3 –Metabolism 13 3No matchb 2 –Otherc 10 8Pathogenesis related 15 –Phenylpropanoid pathway 5 4Photosynthesis related – 30Known plant defense related 14 1Ribosomal protein 4 –Senescence associated 2 1Signaling 8 1Stress related 14 3Transcription factor 9 1Transport 5 1Ubiquitination pathway 4 1Unknownd 12 3Water transport 4 –

Total 153 70a Up- or down-regulated in R11-12 compared with R11-

13. b Genes that have no homology to an annotated gene se-quence in the database. c Genes that encode proteins of othermiscellaneous functions. d Genes that encode proteins of un-known function.




Table II. Genes that are induced during the Pto-mediated plant defense response and also during human and/or fruitfly immune response

�, Differential expression; ∧ , mouse immune response.

GenBankAccession

No.

No. of SimilarGenes Up-Regulated

during PlantDefense

Gene (Best Blast Hit)Functional

Classification

HumanImmune

Responsea

FruitflyImmune

ResponseaReferences

AI779773 1 Phospholipid hydroperoxide glutathioneperoxidase (tomato, Lycopersicon escu-lentum)

Cell protectant � Beck (2001)

AI489449 1 Catalase (tomato) Cell protectant � Zancope-Oliver et al. (1999)AI772349 1 Ferritin tobacco (Nicotiana tabacum) Cell protectant � Huang et al. (2001)AI774583 6 GST (Solanum commersonii) Cell protectant � � Huang et al. (2001); Irving et al. (2001)AI775811 2 MT-like protein (tomato) Cell protectant � � Huang et al. (2001); Irving et al. (2001)CB751576 1 NADP-dependent isocitrate dehydrogenase-

like protein (tomato)Cell protectant � Jo et al. (2001)

CB751574 1 Quinone-oxidoreductase QR1 (Triphysariaversicolor)

Cell protectant � Ross et al. (2000)

AI488458 1 Superoxide dismutase (Cu-Zn SOD; tomato) Cell protectant � Huang et al. (2001)AI776377 2 Thioredoxin, protein disulfide-isomerase

precursor (tobacco)Cell protectant � � Huang et al. (2001); Irving et al. (2001)

AI777808 3 Lipid transfer protein LTP1 precursor (Capsi-cum annuum)

Plant defense/cellwall associated

� Irving et al. (2001)

AI485587 1 Arg decarboxylase 1 (Datura stramonium) Metabolism � Efron and Barbul (1999)CB751579 1 Fructose-1,6-bisphosphatase precursor (pota-

to, Solanum tuberosum)Metabolism � Mabondzo et al. (1991)

AI778001 1 Glu synthetase/glutamate-ammonia ligase(Nicotiana plumbaginifolia)

Metabolism � Karinch et al. (2001)

AI781779 1 S-adenosylmethionine decarboxylase proen-zyme (potato)

Metabolism � Di Leo et al. (1999)

AI772723 1 Thiamin biosynthetic enzyme (soybean, Gly-cine max)

Metabolism � Axelrod (1981)

AI488639 1 UDP-galactose/Glc 4-epimerase-like protein(Arabidopsis)

Metabolism � Gregorio et al. (2001)

CB751578 1 Uricase/urate oxidase (Lotus japonicus) Metabolism � Gregorio et al. (2001); Irving et al.(2001)

CB751580 1 Actin (Malva pusilla) Other � Irving et al. (2001)CB751550 1 DNA repair protein RAD23 (tomato and

fruitfly)Other � Brocksted et al. (1998)

AI775854 1 Ferredoxin-nitrite reductase (tobacco) Other � Gregorio et al. (2001)AI488991 1 Histone H2A (Euphorbia esula) Other � Huang et al. (2001)CB751552 1 Vacuolar proton-inorganic pyrophosphatase

(tobacco)Other � Komissarenko et al. (1986)

AI484482 5 Chitinase (tomato) Pathogenesis-related

� Irving et al. (2001)

CB751582 1 Cytochrome P450 monooxygenase (tobacco) Phenylpropanoidpathway

� � Huang et al. (2001); Irving et al. (2001)

AW032166 1 Hin1 (tobacco), putative syntaxin(Arabidopsis)

Known plant de-fense related

� Shukla et al. (2000)

CB751584 1 Leu aminopeptidase (tomato) Known plant de-fense related

� Gregorio et al. (2001); Irving et al.(2001)

CB751558 1 Lipoxygenase (tomato) Known plant de-fense related

� Malaviya and Abraham (2001)

AI774536 4 Ribosomal protein (N. plumbaginifolia) Ribosomal protein � � Huang et al. (2001); Irving et al. (2001)CB751586 1 Aconitate hydratase or aconitase (potato) Signaling �∧ Pitarch et al. (2001)AI782784 2 Heat shock protein (tomato) Stress related � � Huang et al. (2001); Gregorio et al.

(2001)AI777200 1 Putative trypsin inhibitor (Arabidopsis);

Lemir (tomato)Stress related � Gregorio et al. (2001)

BG625896 3 Putative alcohol dehydrogenase (tomato) Stress/plant de-fense related

�∧ � Irving et al. (2001); Pitarch et al. (2001)

CB751569 1 ABC1 protein (N. plumbaginifolia); PDR5-like ABC transporter

Transport � Irving et al. (2001)

CB751568 1 ADP-ribosylation factor (Arabidopsis) Transport � Gregorio et al. (2001)(Table Continues)

Mysore et al.



R11-12 plants without any pathogen inoculation. Inaddition to these 62 genes, other genes encodingtranscription factors, kinases, transporters, and pro-teins involved in general metabolism were com-monly differentially expressed between plants, hu-mans, and fruitfly defense responses. We did notcompare the 70 genes that were less abundant inR11-12 (when compared with R11-13) with humanand fruitfly defense responses because the majorityof them encode photosynthesis- and chloroplast-associated proteins. We also compared our previ-ously identified differentially expressed genes (My-sore et al., 2002) at a very early time (4 h) afterinfection from RG-PtoR plants inoculated with P.syringae pv tomato(avrPto) with immune responsegenes from human and fruitfly. Interestingly, veryearly after infection, only 23 genes of 227 inducedgenes (approximately 10%) in RG-PtoR were similarto human and/or fruitfly immune response genes(data not shown).

DISCUSSION

Signaling pathways leading to activation of de-fense response genes in mammals, insects, and plantsshare many similar components (Cohn et al., 2001;Staskawicz et al., 2001). For example, the resistanceprotein Pto shares sequence similarity with the IRAKproteins in human and Pelle in fruitfly, both of whichplay central roles in controlling the innate immuneresponse (Cohn et al., 2001). To investigate whether“downstream” defense responses are similar amongmammals, insects, and plants, we compared the PORgenes with previously published human and fruitflyimmune response genes (Gregorio et al., 2001; Huanget al., 2001; Irving et al., 2001). Strikingly, 40% of thePOR genes that were abundantly expressed in Pto-overexpressing plants are functionally similar togenes previously shown to be induced during humanand/or fruitfly immune responses (Table II). Some ofthese could be general stress-responsive genes and

may not be specific to pathogen-induced immuneresponse. However, they may also still play a role inconferring immunity against pathogens. These re-sults suggest that defense responses in these diverseorganisms are either conserved or may have arisenby convergent evolution. For example, genes encod-ing metallothionein (MT), GST, thioredoxin (TRX),cytochrome P450 (P450), heat shock proteins (HSPs),ribosomal proteins, and ubiquitin-conjugating en-zymes are all induced during defense responses inPto-overexpressing plants, humans, and fruitfly (Ta-ble II). Although these genes are reasonably wellcharacterized in mammalian systems, their actualrole in immunity is not clear. Because of the avail-ability of high-throughput techniques like virus-induced gene silencing and reverse genetics inplants, it might be easier to investigate the function ofthese in immunity by using a plant model system.

The oxidative burst that precedes HR, a commonphenomenon triggered by an incompatible plantpathogen interaction, has been suggested to be aprimary event responsible for triggering the cascadeof defense responses in various plant species againstinfection with avirulent pathogens or pathogen-derived elicitors. The oxidative burst may be fol-lowed by activation of genes encoding antioxidantenzymes in tissue surrounding the initial infectionsite (Lamb and Dixon, 1997) to protect the surround-ing cells against oxidative damage. We found thatseveral of these cell protectant genes (POR1–POR19),like glutathione peroxidase, separation anxiety pro-tein, catalase, superoxide dismutase, ferritin, GST,MT, NADP-dependent isocitrate dehydrogenase,ascorbate oxidase, quinone-oxidoreductase, andTRX, were up-regulated during the Pto-mediated dis-ease resistance response. A majority of these wasconstitutively up-regulated in R11-12 plants withoutany pathogen infection. Interestingly, all these genes,with the exception of separation anxiety protein andascorbate oxidase, were also shown to be inducedduring human immune response (Table II). This sug-

Table II. Continued

GenBankAccession

No.

No. of SimilarGenes Up-Regulated

during PlantDefense

Gene (Best Blast Hit)Functional

Classification

HumanImmune

Responsea

FruitflyImmune

ResponseaReferences

AI485102 1 Synaptobrevin-like V snare protein (yeast,Saccharomyces cerevisiae, and human,Homo sapiens)


AI775530 4 Water channel protein (Nicotiana excelsior);aquaporin-like protein (Petunia �

hybrida)


AI488816 3 E2 ubiquitin-conjugating-like enzyme Ahus5(Arabidopsis)

Ubiquitinationpathway

� � Huang et al. (2001); Irving et al. (2001)

CB751590 1 Polyubiquitin (maize, Zea mays) Ubiquitinationpathway

� Ishii et al. (1999)

a Similarity is inferred by putative function and not by sequence relationships.




gests that the cell protectants that are induced duringan immune response are highly conserved amonganimals and plants.

MTs and GSTs are a group of stress and immuneresponse proteins that contribute to cellular survivaldue to oxidative damage (Borghesi and Lynes, 1996).TRX plays multiple roles in intracellular signalingand resistance against oxidative stress (Hirota et al.,2002). In humans, the TRX gene contains a series ofstress-responsive elements; in turn, TRX promotesactivation of transcription factors such as NF-kappaB, AP-1, and p53 that are components of the humanimmune system (Hirota et al., 2002). In vertebrates,the wide variety of P450s is a key to many of theknown defense mechanisms of vertebrates against allxenobiotic forms (Morfin, 2002). Taken together,overexpression of MTs, GSTs, TRX, and P450s duringimmune responses in humans, insects, and plantsseems to participate in the same basic function of cellprotection. HSPs act as chaperones, are widely dis-tributed in nature, and are among the most highlyconserved molecules of the biosphere (Zugel andKaufmann, 1999). HSP synthesis is increased to pro-tect prokaryotic or eukaryotic cells from various in-sults during periods of stress caused by infection,inflammation, or similar events (Zugel and Kauf-mann, 1999). Reversal of polypeptide unfolding andpreventing protein aggregation are major functionsof HSPs, especially under stress, and they seem toperform the same functions in animals, plants, andinsects (Zugel and Kaufmann, 1999).

Protein translation plays an important role duringinnate immunity in plants, animals, and insects. Forexample, the fruitfly Thor gene has been shown to berequired for innate immunity in fruitfly by preferen-tial translation of immune transcripts (Bernal andKimbrell, 2000). During the immune and defenseresponses in mammals, insects, and plants, massiveproduction of antimicrobial proteins is required, andtranslation of mRNAs encoding proteins with coun-teractive effects must be suppressed. Hence, modu-lation of ribosomal gene expression during immuneresponses in humans, fruitfly and plants is probablyessential.

Protein ubiquitination is not only essential for thenormal physiological turnover of proteins but ap-pears to have been adapted as part of an intracellularsurveillance system that can be activated by altered,damaged, or foreign proteins and organelles and hasevolved to be an “intracellular immune system”(Ben-Neriah, 2002). Ubiquitination pathway geneshave been shown to be induced during both humanand fruitfly immune response (Huang et al., 2001;Irving et al., 2001; Table II). Recently, it has beenshown that mutations in a plant homolog of animalSGT1, a component of ubiquitin ligase, disable earlydefenses conferred by multiple R genes (Austin et al.,2002; Azevedo et al., 2002). SGT1 is also shown toplay an important role in N gene-mediated resistance

response to tobacco mosaic virus (Liu et al., 2002b).In this study, we have shown that four ubiquitinationpathway-related genes (POR200 and POR202-204)were induced during Pto-mediated disease resistanceresponse. These data suggest that the ubiquitination-mediated protein degradation pathway plays an im-portant role in both plant and animal defense.

Pto-overexpressing plants (R11-12) are slightlystunted, and they develop small white veins in theleaves as they age (Li et al., 2002). It is possible thatthese phenotypes are due to pleiotropic effects of Ptooverexpression that are unrelated to the defense re-sponse. However, it is unlikely that differential ex-pression of most of the POR genes is due to pleiotropiceffects because we found that the majority of the genesare differentially expressed upon inoculation by P.syringae pv tomato(avrPto), and they fall into the samefunctional classes that were shown to be differentiallyexpressed during a disease resistance response in RG-PtoR plants (Mysore et al., 2002; see supplementaldata Table S1 at http://www.plantphysiol.org).

Requirement of Prf for the constitutive plant de-fense response mediated by Pto overexpression (Xiaoet al., 2003) suggests that both R11-12 and RG-PtoRplants initiate similar defense responses in responseto P. syringae pv tomato expressing avrPto. Eventhough the POR genes belong to the same functionalclasses identified in this earlier study, approximately80% represent previously undiscovered genes. Thislack of significant overlap could be due to severalreasons. First, the expression profiling we present inthis paper is likely not as comprehensive as thatreported earlier; further overlap of the genes mightbe revealed with a more in-depth profiling approachsuch as GeneCalling (Mysore et al., 2002). Second, itseems likely that Pto overexpression “amplifies” thegene expression differences that normally occur inplants expressing Pto from its native promoter. Thisexaggerated response might have allowed the iden-tification of genes in the Pto-overexpressing line thatare normally expressed at much lower levels duringthe “natural” Pto-mediated resistance response.Third, the differentially expressed genes identified inthis study were from tissue collected at 4 and 8 h afterinfection, whereas in the previous study it was onlyfrom tissue collected at 4 h after infection. Fourth, arecent report suggests that the AvrPto-independentdefense activation in R11-12 and gene-for-gene resis-tance mediated by Pto are functionally separable.Even though Pto-mediated gene-for-gene resistanceand AvrPto-independent resistance require both Ptoand Prf, it was recently shown that AvrPto-independent resistance appears to be distinct fromgene-for-gene resistance in terms of the requirementof downstream components, sensitivity to mutationsin Pto, and defense activation (Xiao et al., 2003).

A recent report from Xiao et al. (2001) providedsome insights about differential gene expression inthe Pto-overexpressing line, and we have now signif-

Mysore et al.



icantly expanded on that study. Our study differsfrom that of Xiao et al., first, by investigating not onlygenes that are differentially expressed withoutpathogen inoculation in the Pto-overexpressing linebut also those genes that are induced in this lineupon inoculation by P. syringae pv tomato(avrPto).Second, we also identified a set of genes that aresuppressed in the Pto-overexpressing plants. Third,because results are now available from a comprehen-sive transcript profiling experiment of a tomato lineexpressing Pto from its native promoter, we wereable to compare the POR genes with those differen-tially expressed during the “normal” resistance re-sponse. Fourth, we examined the expression of thePOR genes over a time course, and this allowed us tocluster them based on their expression patterns. Wedid observe some overlap of the POR genes withthose identified by Xiao et al. Approximately 52% ofthe genes identified by Xiao et al. are identical to ourPOR genes. Together, these two studies have identi-fied a total of 255 genes that are differentially ex-pressed in the Pto-overexpressing lines.

We identified 21 different classes of genes thatwere differentially expressed in R11-12 after inocula-tion with P. syringae pv tomato expressing avrPto (Ta-ble I). Interestingly, genes belonging to these classes,with the exception of chloroplast- or photosynthesis-related genes, water transport-related genes, andubiquitination pathway genes, were constitutivelyexpressed in the R11-12 plants in the absence of P.

syringae pv tomato expressing avrPto. Figure 4 depictsa model for constitutive plant defense responses dueto Pto overexpression.

Phenylpropanoids have been proposed to serve asflower pigments (anthocyanin), UV protectants, de-fense chemicals (phytoalexins and insect repellents),allelopathic agents, and signal molecules in plant-microbe interactions. We identified 10 different genes(POR97-105) involved in the phenylpropanoid path-way that are differentially expressed in the Pto-overexpressing plants, R11-12. Five of these 10 geneswere constitutively differentially expressed in R11-12plants. Cell wall reinforcement and thickening areassociated with plant defense during resistance re-sponse. In this study, we have shown that majority ofthe cell wall-associated genes (POR20-21, POR23-24,and POR26-27) were constitutively up-regulated inR11-12 plants.

Overlap between the leaf senescence and pathogendefense programs has been reported earlier (Quirinoet al., 1999). Obvious visual symptoms of leaf senes-cence are loss of chlorophyll pigments due to declinein photosynthesis and chloroplast disintegration(Buchanan-Wollaston, 1997). It has been suggestedthat in cells undergoing HR, a decline in photosyn-thesis might act as an inducer of senescence (Quirinoet al., 2000). In this study, we show 30 photosyn-thesis-related genes (POR106–135) that were sup-pressed during the Pto-mediated disease resistanceresponse. We also identified 12 genes encoding chlo-

Figure 4. A summary of the plant responses tooverexpression of Pto and P. syringae pv tomatoinfection. Red arrows indicate genes that wereinduced and green arrows indicate genes thatwere suppressed due to Pto overexpression.Black arrows indicate either possible steps insignal transduction pathways or host responsesthat might directly impact other responses.Classes of genes shaded in gray are also respon-sive to P. syringae pv tomato infection.




roplast associated proteins (POR28–39) that were alsosuppressed during the resistance response. Interest-ingly, pathogen inoculation was required for the sup-pression of these genes. These observations suggestthat, in the face of pathogen attack, the plant is rap-idly redirecting its resources from growth processesto defense responses.

Water is essential for pathogen multiplication inthe plant tissue. One of the P. syringae pv tomatodisease symptoms is water-soaked lesions on tomatoleaves. In this study, we have shown that four aqua-porin/tonoplast intrinsic proteins (POR220-223)were induced during the resistance response. Inter-estingly, aquaporin genes were suppressed during apreviously studied Pto (native promoter driven)-mediated disease resistance response (Mysore et al.,2002). It seems like the Pto overexpression is slightlyup-regulating the aquaporin genes, and its expres-sion is enhanced after pathogen infection. The actualrole of water transport during plant defense responseis yet to be determined.

Several genes encoding known components of thePto signal transduction pathway were deliberately in-cluded in the microarray experiments to study theirexpression in Pto-overexpressing plants (see supple-mental data Table S1 at http://www.plantphysiol.org). As expected, the Pto gene (POR161) was abun-dantly expressed in the R11-12 plants when comparedwith the azygous plants at the 0-h time point. Pto wasslightly induced due to P. syringae pv tomato express-ing avrPto inoculation. This was probably due to theCaMV 35S promoter used to overexpress the Ptogene. Tomato plants infected by the avirulent P. sy-ringae pv tomato have been shown to accumulate SA(Oldroyd and Staskawicz, 1998), and the CaMV 35Spromoter is responsive to SA (Qin et al., 1994). Pti4,Pti5, and Pti6 are ERF transcription factors that inter-act with Pto; expression of the Pti4/5/6 genes isinduced in both the compatible and incompatibleinteractions, whereas the induction of Pti5 is furtherenhanced during incompatible interaction whencompared with compatible interaction (Zhou et al.,1997; Thara et al., 1999; Gu et al., 2000). Here, wehave shown that genes encoding Pti4 (POR186) andPti5 (POR185) are constitutively and abundantly ex-pressed in R11-12 plants without any pathogen inoc-ulation. Previously, using RNA-blot hybridizations,Pti4 was shown to be equally expressed in both Pto-overexpressing and azygous lines (Tang et al., 1999;Gu et al., 2000). This discrepancy is most likely due tothe differences in the sensitivity of RNA-blot hybrid-ization and microarray hybridization. It is interestingto note that Pti4 was also induced in the RG-PtoRplants, in response to P. syringae pv tomato expressingavrPto, when compared with Rio-Grande tomatoplants without Pto (Mysore et al., 2002). Pti6 was notdifferentially expressed due to Pto overexpression(data not shown).

Api1 and Api2 encode tomato proteins that interactwith AvrPto in a yeast two-hybrid system (Bog-danove and Martin, 2000). Using microarrays, wehave shown that Api1 (POR139) was suppressed inR11-12 plants at 4 and 8 h after inoculation with P.syringae pv tomato expressing avrPto. Api1 encodes astress-related protein that is induced by heavy metal,wounding, and virus inoculation. It is possible thatApi1 is a virulence target for AvrPto, and, hence, theplant is suppressing the gene expression as a defensemechanism. Api2 (POR42), which encodes a Ras-related Rab8-like protein, was constitutively andabundantly expressed in R11-12 plants without anypathogen inoculation. Api2 is involved in vesicularprotein trafficking, and its role in disease formationis yet to be determined. Adi1 encodes an AvrPto-dependent Pto-interacting tomato protein (Bog-danove and Martin, 2000). Adi1 is a catalase that isinvolved in scavenging hydrogen peroxide (Bog-danove and Martin, 2000). In this study, we haveidentified a tomato catalase (POR7) that is constitu-tively abundantly expressed in the R11-12 plantswithout any pathogen inoculation. This catalase ismost likely involved in protecting the cells from ex-cessive damage due to oxidative burst that resultsfrom Pto overexpression.

The possible role of Pto in a basal defense responseis suggested by our recent report that the RG-PtoRplants constitutively induce several defense-relatedgenes, including PR genes without any pathogen in-fection (Mysore et al., 2002). When Pto is overex-pressed, such basal defense response is exaggerated toup-regulate an array of defense-related genes (PORgenes). Constitutive overexpression of ERF transcrip-tion factors Pti4 and Pti5 (Zhou et al., 1997), withoutany pathogen infection, in Pto-overexpressing plantsfurther corroborates a role for Pto in basal defense andsupports the notion that Pto can activate Pti4 and Pti5in the absence of AvrPto (Zhou et al., 1997). Interest-ingly, many of the POR genes are also known to beinduced during immune responses in human and/orfruitfly. This observation further supports the notionthat defense responses among plants, mammals, andinsects have been conserved during evolution. Futurecharacterization of the POR genes by virus-inducedgene silencing (Liu et al., 2002a) and other loss- andgain-of-function methods should help to clarify thepotential role of these genes in disease resistance.

MATERIALS AND METHODS

Plant Materials, Bacterial Strains, Plant Inoculation, andRNA Isolation

Tomato (Lycopersicon esculentum) Pto-transgenic line R11-12 (35S::Pto/35S::Pto) and a sibling line R11-13 that has segregated away the Pto trans-gene were described previously (Tang et al., 1999). Seeds of these twogenotypes were germinated on moist Whatman paper (Whatman, Clifton,NJ) and grown in soil in the greenhouse at 26°C to 28°C and 16 h of light perday. Bacterial inoculations were performed as described previously (Gu etal., 2000) and below under cDNA library construction. For SSH, cDNA

Mysore et al.



microarray, and RNA-blot experiments, plants were inoculated with bacte-ria by vacuum infiltration at a concentration of 2.2 � 107 colony-formingunits (cfu) mL�1. RNA was isolated using the hot phenol method as de-scribed previously (Gu et al., 2000).

Construction of cDNA Libraries and EST Sequencing

To create cDNA libraries, 4-week-old plants of tomato lines R11-12 andR11-13 were vacuum infiltrated with Pseudomonas syringae pv tomato strainT1 expressing avrPto at concentrations of either 105 or 108 cfu mL�1. At the108 cfu mL�1 (high titer) inoculation level, leaves were harvested at 0, 2, 4,6, and 8 h after inoculation. Effectiveness of the high-titer inoculations wasverified by assessing defense-related gene expression (in R11-13) or byobservation of the HR (in R11-12). At the 105 cfu mL�1 (low titer) inocula-tion level, leaves were harvested at 0, 12, 24, 36, and 48 h after inoculation.Effectiveness of the low-titer inoculations was verified by assessing defense-related gene expression (in both R11-12 and R11-13). Equal amounts of leaftissues from the R11-12 and R11-13 plants from the different time points ofhigh- and low-titer inoculations for each line were pooled and used toextract total RNA.

Poly(A�) RNA was purified using the poly(A�) tract mRNA isolationsystem (Promega, Madison, WI). An oriented Uni-ZAP XR library wasprepared using a ZAP-cDNA synthesis kit (Stratagene, La Jolla, CA) accord-ing to the manufacturer’s instructions. The cDNA (0.4 �g) was ligated intoUni-ZAP XR digested with XhoI and EcoRI. This yielded 1.4 � 106 and 1.2 �106 primary plaques for the R11-12 (R) library and the R11-13 (S) library. Massexcision was performed on 1 � 107 plaque-forming units from each library.Escherichia coli strain XL1-Blue MRF was used for propagating the library. Thebacterial cultures were subsequently arrayed into 384-well plates and used forsequencing. Additional information about the libraries can be found online(http://sgn.cornell.edu or http://www.tigr.org/tdb/tgi/lgi). Approxi-mately 5,500 clones from each of the R and S libraries were sequenced by TheInstitute for Genomics Research (TIGR; http://www.tigr.org/tdb/tgi/lgi).

In Silico EST Subtraction and cDNA Subtraction by SSH

For in silico EST subtraction (Stekel et al., 2000), 10,872 EST sequencesfrom the R and S libraries were downloaded from the TIGR database andassembled using “TIGR Assembler” running on a Linux operating system.Custom scripts were written in PERL to index each contig resulting from theassembly. The indexing process calculated the number of ESTs within eachcontig derived from the R or S libraries. Contigs that contained more than2-fold the number of ESTs from one library than the other were flagged forfurther analysis (see “Results”).

SSH was done by using the CLONTECH PCR-Select cDNA SubtractionKit (CLONTECH Laboratories Inc., Palo Alto, CA). Equal amounts of RNAfrom leaves 4 and 8 h after inoculation with P. syringae pv tomato(avrPto)were pooled for each of the R11-12 and R11-13 plants and used for SSH. Bothforward (R11-12 as tester and R11-13 as driver) and reverse (R11-13 as testerand R11-12 as driver) subtractions were performed according to the manu-facturer’s protocol. The final PCR products (enriched for cDNAs corre-sponding to differentially expressed transcripts) resulting from the SSHwere either digested with NotI and directly cloned into pBluescript vector orlabeled with radioactive P32 (DECAprime DNA labeling kit, Ambion Inc.,Austin, TX) and used as a probe on colony blots containing a nonredundantEST collection from the R and S libraries. PCR products cloned into pBlue-script vector were sequenced (65 each from forward and reverse subtrac-tions), and the sequences were searched against the NCBI databases. ESTsequences corresponding to colonies that hybridized to the radiolabeledPCR product were obtained from the TIGR databases and were subse-quently used for searches of the NCBI databases.

Microarray Preparation, Hybridization, andData Analysis

Inserts from approximately 650 cDNA clones corresponding to ESTs(http://www.tigr.org) were PCR amplified using T3 and T7 primers. Thisnumber included about 600 ESTs that corresponded to genes obtained bySSH (301) and in silico EST subtraction (299), 38 ESTs that corresponded toknown defense response genes, and 12 controls (e.g. alpha-tubulin, beta-tubulin, and genes encoding ribosomal proteins). PCR products were puri-

fied using MagBeads (Bangs Lab., Fishers, IN). Purified PCR products werevacuum dried and resuspended in 30 �l of spotting buffer (3� SSC � 0.1%[w/v] Sarkosyl). DNA spotting, labeling, hybridization, and data analyseswere performed as described earlier (Mysore et al., 2002). Changes in mRNAexpression in excess of 2-fold between the two samples were considered forthese experiments. Cluster and Treeview software (Eisen et al., 1998) wereused (http://rana.lbl.gov/EisenSoftware.htm) to group and display geneswith similar expression profiles. We used the default options of hierarchicalclustering.

Several quality control measures were implemented in our analysis ofthe microarray data to meet the Minimum Information About a Micro-array Experiment (Brazma et al., 2001) guidelines (http://www.mged.org/Workgroups/MIAME/miame.html). The variation in abundance (co-efficient of variation shown in supplementary Table S1 at http://www.plantphysiol.org) for each transcript was estimated using two biologicalreplications and at least two technical replications for each biological rep-lication. PCR products corresponding to each cDNA were replicated threetimes on the same slide to control for possible variations in hybridizationwithin a slide. The ratios shown in Table S1 (http://www.plantphysiol.org)are averages of up to 12 independent ratios representing each gene (twobiological replications � two technical replications � three spots per repli-cate). The correlation coefficient between biological replicates was 0.812 andbetween technical replicates was 0.923. By dye swapping, we showed thatthere was no significant variation due to possible biases in direct labelingmethod of Cy3 or Cy5 used in this study. The correlation coefficient forthe dye-swap experiment was 0.918. All the identified POR genes wereresequenced from both 3� and 5� ends to confirm their gene identity. Toconform with Minimum Information About a Microarray Experiment guide-lines, we have provided the raw data (before normalization) of spot inten-sities and an example of an overlay image in supplementary Table S2 andFig. S3 (http://www.plantphysiol.org), respectively.

ACKNOWLEDGMENTS

We thank Paul Debbie (Boyce Thompson Institute Center for Gene Ex-pression Profiling) for technical help with microarray experiments andDavid Wendell for deriving nonredundant cDNAs.

Received February 26, 2003; returned for revision March 26, 2003; acceptedMay 3, 2003.

LITERATURE CITED

Austin MJ, Muskett P, Kahn K, Feys BJ, Jones JDG, Parker JE (2002)Regulatory role of SGT1 in Early R gene-mediated plant defenses. Science295: 2077–2080

Axelrod AE (1981) Role of the B vitamins in the immune response. Adv ExpMed Biol 135: 93–106

Azevedo C, Sadanandom A, Kitagawa K, Freialdenhoven A, Shirasu K,Schulze-Lefert P (2002) The RAR1 interactor SGT1, an essential compo-nent of R gene-triggered disease resistance. Science 295: 2073–2076

Beck MA (2001) Antioxidants and viral infections: host immune responseand viral pathogenicity. J Am Coll Nutr 20: 384S–388S

Ben-Neriah Y (2002) Regulatory functions of ubiquitination in the immunesystem. Nat Immunol 3: 20–26

Bernal A, Kimbrell DA (2000) Drosophila Thor participates in host immunedefense and connects a translational regulator with innate immunity.Proc Natl Acad Sci USA 97: 6019–6024

Bogdanove AJ, Martin GB (2000) AvrPto-dependent Pto-interacting pro-teins and AvrPto-interacting proteins in tomato. Proc Natl Acad Sci USA97: 8836–8840

Borghesi LA, Lynes MA (1996) Stress proteins as agents of immunologicalchange: some lessons from metallothionein. Cell Stress Chaperones 1:99–108

Brazma A, Hingamp P, Quackenbush J, Sherlock G, Spellman P, StoeckertC, Aach J, Ansorge W, Ball CA, Causton HC et al. (2001) Minimuminformation about a microarray experiment (MIAME)-toward standardsfor microarray data. Nat Genet 29: 365–371

Brocksted E, Rickers A, Kostka S, Laubersheimer A, Dorken B, Wittmann-Liebold B, Bommert K, Otto A (1998) Identification of apoptosis-associated proteins in a human Burkitt lymphoma cell line: cleavage of




heterogeneous nuclear ribonucleoprotein A1 by caspase 3. J Biol Chem273: 28057–28064

Buchanan-Wollaston V (1997) The molecular biology of leaf senescence. JExp Bot 48: 181–199

Cohn J, Sessa G, Martin GB (2001) Innate immunity in plants. Curr OpinImmunol 13: 55–62

Di Leo A, Messa C, Russo F, Linsalata M, Amati L, Caradonna L, Pece S,Pellegrino NM, Caccavo D, Antonaci S et al. (1999) Helicobacter pyloriinfection and host cell responses. Immunopharmacol Immunotoxicol 21:803–846

Diatchenko L, Lau Y-FC, Campbell AP, Chenchik A, Moqadam F, HuangB, Lukyanov S, Lukyanov K, Gurskaya N, Sverdlov ED et al. (1996)Suppression subtractive hybridization: a method for generating differen-tially regulated or tissue-specific cDNA probes and libraries. Proc NatlAcad Sci USA 93: 6025–6030

Efron DT, Barbul A (1999) Arginine and nutrition in renal disease. J RenNutr 9: 142–144

Eisen MB, Spellman PT, Brown PO, Botstein D (1998) Cluster analysis anddisplay of genome-wide expression patterns. Proc Natl Acad Sci USA 95:14863–14868

Galan JE, Collmer A (1999) Type III secretion machines: bacterial devicesfor protein delivery into host cells. Science 284: 1322–1328

Gregorio ED, Spellman PT, Rubin GM, Lemaitre B (2001) Genome-wideanalysis of the Drosophila immune response by using oligonucleotidemicroarrays. Proc Natl Acad Sci USA 98: 12590–12595

Gu YQ, Yang C, Venkatappa TK, Zhou J, Martin GB (2000) Pti4 is inducedby ethylene and salicylic acid, and its product is phosphorylated by thePto kinase. Plant Cell 12: 771–785

He SY (1998) Type III protein secretion systems in plant and animal patho-genic bacteria. Annu Rev Phytopathol 36: 363–392

Hirota K, Nakamura H, Masutani H, Yodoi J (2002) Thioredoxin superfam-ily and thioredoxin-inducing agents. Ann N Y Acad Sci 957: 189–199

Huang Q, Liu D, Majewski P, Schulte LC, Korn JM, Young RA, Lander ES,Hacohen N (2001) The plasticity of dendritic cell responses to pathogensand their components. Science 294: 870–875

Irving P, Troxler L, Heuer TS, Belvin M, Kopczynski C, Reichhart JM,Hoffman JA, Hetru C (2001) A genome-wide analysis of immune re-sponses in Drosophila. Proc Natl Acad Sci USA 98: 15119–15124

Ishii T, Itoh K, Sato H, Bannai S (1999) Oxidative stress-inducible proteinsin macrophages. Free Radic Res 31: 351–355

Jo SH, Son MK, Koh HJ, Lee SM, Song IH, Kim YO, Lee YS, Jeong KS,Kim WB, Park JW et al. (2001) Control of mitochondrial redox balanceand cellular defense against oxidative damage by mitochondrialNADP�-dependent isocitrate dehydrogenase. J Biol Chem 276:16168–16176

Karinch AM, Pan M, Lin CM, Strange R, Souba WW (2001) Glutaminemetabolism in sepsis and infection. J Nutr 131: 2535S–2538S

Kim YJ, Lin N-C, Martin GB (2002) Two distinct Pseudomonas effectorproteins interact with the Pto kinase and activate plant immunity. Cell109: 589–598

Komissarenko SV, Gulaia NM, Gaivoronskaia GG, Karlova NP, TarusovaNB (1986) Inorganic pyrophosphatase activity of the mouse spleen in theimmune response and after treatment with bis-phosphonates. Ukr Bio-khim Zh 58: 22–27

Lamb C, Dixon RA (1997) The oxidative burst in plant disease resistance.Annu Rev Plant Physiol Plant Mol Biol 48: 251–275

Li J, Shan L, Zhou J-M, Tang X (2002) Overexpression of Pto induces asalicylate-independent cell death but inhibits necrotic lesions caused bysalicylate-deficiency in tomato plants. Mol Plant-Microbe Interact 15:654–661

Liu Y, Schiff M, Dinesh-Kumar SP (2002a) Virus-induced gene silencing intomato. Plant J 31: 777–786

Liu Y, Schiff M, Serino G, Deng X-W, Dinesh-Kumar SP (2002b) Role ofSCF ubiquitin-ligase and the COP9 signalosome in the N gene mediatedresistance response to Tobacco mosaic virus. Plant Cell 14: 1483–1496

Mabondzo A, Le Naour R, Raoul H, Clayette P, Lafuma C, Barre-SinoussiFC, Cayre Y, Dormont D (1991) In vitro infection of macrophages byHIV: correlation with cellular activation, synthesis of tumour necrosisfactor alpha and proteolytic activity. Res Virol 142: 205–212

Malaviya R, Abraham SN (2001) Mast cell modulation of immune re-sponses to bacteria. Immunol Rev 179: 16–24

Martin GB, Brommonschenkel SH, Chunwongse J, Frary A, Ganal MW,Spivey R, Wu T, Earle ED, Tanksley SD (1993) Map-based cloning of aprotein kinase gene conferring disease resistance in tomato. Science 262:1432–1436

Morfin R (2002) Involvement of steroids and cytochrome P(450) species inthe triggering of immune defenses. J Steroid Biochem Mol Biol 80:273–290

Mysore KS, Crasta OR, Tuori RP, Folkerts O, Swirsky PB, Martin GB(2002) Comprehensive transcript profiling of Pto- and Prf-mediated hostdefense responses to infection by Pseudomonas syringae pv. tomato. Plant J32: 299–315

Nurnberger T, Brunner F (2002) Innate immunity in plants and animals:emerging parallels between the recognition of general elicitors andpathogen-associated molecular patterns. Curr Opin Plant Biol 5: 318–324

Ogura Y, Bonen DK, Inohara N, Nicolae DL, Chen FF, Ramos R, Britton H,Moran T, Karaliuskas R, Duerr RH et al. (2001a) A frameshift mutationin NOD2 associated with susceptibility to Crohn’s disease. Nature 411:603–606

Ogura Y, Inohara N, Benito A, Chen FF, Yamaoka S, Nunez G (2001b)Nod2, a Nod1/Apaf-1 family member that is restricted to monocytes andactivates NF-kB. J Biol Chem 276: 4812–4818

Oldroyd GED, Staskawicz BJ (1998) Genetically engineered broad-spectrum disease resistance in tomato. Proc Natl Acad Sci USA 95:10300–10305

Pitarch A, Diez-Orejas R, Molero G, Pardo M, Sanchez M, Gil C, NombelaC (2001) Analysis of the serological response to systematic Candida albi-cans infection in a murine model. Proteomics 1: 550–559

Qin XF, Holuigue L, Horvath DM, Chua NH (1994) Immediate earlytranscription activation by salicylic acid via the cauliflower mosaic virusas-1 element. Plant Cell 6: 863–874

Quirino BF, Noh YS, Himelblau E, Amasino RM (2000) Molecular aspectsof leaf senescence. Trends Plant Sci 5: 278–282

Quirino BF, Normanly J, Amasino RM (1999) Diverse range of gene activityduring Arabidopsis thaliana leaf senescence includes pathogen-independent induction of defense-related genes. Plant Mol Biol 40:267–278

Ross D, Kepa JK, Winski SL, Beall HD, Anwar A, Siegel D (2000) NAD(P)H:quinone oxidoreductase 1 (NQO1): chemoprotection, bioactivation,gene regulation and genetic polymorphisms. Chem Biol Interact 129:77–97

Shukla A, Berglund L, Nielsen LP, Nielsen S, Hoffman HJ, Dahl R (2000)Regulated exocytosis in immune functions: are SNARE-proteins in-volved? Respir Med 94: 10–17

Staskawicz BJ, Mudgett MB, Dangl JL, Galan JE (2001) Common andcontrasting themes of plant and animal diseases. Science 292: 2285–2289

Stekel DJ, Git Y, Falciani F (2000) The comparison of gene expression frommultiple cDNA libraries. Genome Res 10: 2055–2061

Tang X, Xie M, Kim YJ, Zhou J, Klessig DF, Martin GB (1999) Overexpres-sion of Pto activates defense responses and confers broad resistance.Plant Cell 11: 15–30

Thara VK, Tang X, Gu YQ, Martin GB, Zhou JM (1999) Pseudomonas syringaepv tomato induces the expression of tomato EREBP-like genes Pti4 and Pti5independent of ethylene, salicylate and jasmonate. Plant J 20: 475–483

Xiao F, Lu M, Li J, Zhao T, Yi SH, Thara VK, Tang X, Zhou J (2003) Ptomutants differentially activate Prf-dependent, avrPto-independent resis-tance and gene-for-gene resistance. Plant Physiol 131: 1239–1249

Xiao F, Tang X, Zhou J (2001) Expression of 35S::Pto globally activatesdefense-related genes in tomato plants. Plant Physiol 126: 1637–1645

Zancope-Oliver RM, Reiss E, Lott TJ, Mayer LW, Deepe GS (1999) Molec-ular cloning, characterization, and expression of the M antigen of His-toplasma capsulatum. Infect Immunol 67: 1947–1953

Zhou J, Tang X, Martin GB (1997) The Pto kinase conferring resistance totomato bacterial speck disease interacts with proteins that bind a cis-element of pathogenesis-related genes. EMBO J 16: 3207–3218

Zugel U, Kaufmann SHE (1999) Role of heat shock proteins in protectionfrom and pathogenesis of infectious diseases. Clin Microbiol Rev 12:19–39

Mysore et al.



Documents

Overexpression of the Disease Resistance Gene Pto in Tomato Induces Gene … · Overexpression of the Disease Resistance Gene Pto in Tomato Induces Gene Expression Changes Similar