Proc. Natl. Acad. Sci. USAVol. 92, pp. 6597-6601, July 1995Plant Biology

NDRJ, a locus ofArabidopsis thaliana that is required fordisease resistance to both a bacterial and a fungal pathogen

(Pseudomonas syringae/Peronospora parasitica/hypersensitive response/signal transduction)

KAREN S. CENTURY*, ERIC B. HOLUBt, AND BRIAN J. STASKAWICZt§Departments of *Environmental Science, Policy, and Management and tPlant Biology, University of California, Berkeley, CA 94720; and tDepartment of PlantPathology and Weed Science, Horticultural Research International-East Malling, West Malling, Kent, ME19 6BJ, United Kingdom

Communicated by Frederick M. Ausubel, Massachusetts General Hospital, Boston, MA, March 6, 1995 (received for review December 19, 1994)

ABSTRACT We have employed Arabidopsis thaliana as amodel host plant to genetically dissect the molecular pathwaysleading to disease resistance. A. thaliana accession Col-0 issusceptible to the bacterial pathogen Pseudomonas syringae pv.tomato strain DC3000 but resistant in a race-specific mannerto DC3000 carrying any one of the cloned avirulence genesavrB, avrRpml, avrRpt2, and avrPph3. Fast-neutron-muta-genized Col-0 M2 seed was screened to identify mutantssusceptible to DC3000(avrB). Disease assays and analysis ofinplanta bacterial growth identified one mutant, ndrl-l (non-race-specific disease resistance), that was susceptible toDC3000 expressing any one of the four avirulence genes tested.Interestingly, a hypersensitive-like response was still inducedby several of the strains. The ndrl-l mutation also renderedthe plant susceptible to several avirulent isolates of the fungalpathogen Peronospora parasitica. Genetic analysis of ndrl-1demonstrated that the mutation segregated as a single reces-sive locus; located on chromosome III. Characterization of thendrl-1 mutation suggests that a common step exists in path-ways of resistance to two unrelated pathogens.

Genetic studies of plant hosts and their pathogens haverevealed that race-specific disease resistance is often con-trolled by corresponding single genetic loci in the two inter-acting partners (1, 2). Absence or inactivation of either genemember results in host susceptibility. Although the phenotypeof resistance in the plant host may vary depending on theparticular host species and pathogen involved, resistance isoften correlated with a rapid localized necrosis called thehypersensitive response (HR) (3). Several events are associ-ated with the HR, including the accumulation of pathogenesis-related (PR) proteins, production of antimicrobial compounds(phytoalexins), oxidative burst, ion fluxes, and cell wallstrengthening (4, 5). It is not clear which of these responses arerequired for the limitation of pathogen growth.

Progress in understanding the molecular events controllingthe specificity of plant disease resistance has been greatlyadvanced by the molecular cloning of avirulence genes fromthe pathogen and the corresponding resistance genes from thehost (2, 6-11). The recent cloning of several race-specificresistance genes from plants has revealed possible mechanismsused for successful plant defense. The gene product of thetomato gene PTO, conferring race-specific resistance toPseudomonas syringae pv. tomato (Pst), has homology to pro-tein-serine/threonine kinases from various plants, suggestingthat protein phosphorylation is involved in the signal trans-duction pathway leading to the resistance response (7). Thepredicted proteins of resistance genes RPS2 from Arabidopsisthaliana (8, 9),N from tobacco (10), and Cf-9 from tomato (11)all have leucine-rich repeat regions, which are implicated inprotein-protein interactions (reviewed in ref. 12). Sequence

The publication costs of this article were defrayed in part by page chargepayment. This article must therefore be hereby marked "advertisement" inaccordance with 18 U.S.C. §1734 solely to indicate this fact.

homology between RPS2 and N, which confer resistance to a

bacterial and a viral pathogen, respectively, suggests thatmechanisms of resistance are conserved among different plantspecies and in response to vastly different types of pathogens.It is probable that one or more biochemical steps are commonto signal transduction pathways initiated by different pathogenspecies and races within a species.The interactions between A. thaliana and the pathogens

Pseudomonas syringae (a bacterial pathogen) and Peronosporaparasitica (a fungal pathogen) have been used as model systemsto dissect the molecular pathways leading to disease resistance(reviewed in refs. 13 and 14). We have used the interactionbetween A. thaliana and Pst to identify a locus in A. thalianaputatively involved in a shared pathway for disease resistance.We describe here the characterization of this locus, designatedNDRJ (for non-race-specific disease resistance), which is re-

quired for resistance not only to Pst expressing any one ofseveral cloned avirulence genes but also to various races of P.parasitica.

MATERIALS AND METHODS

Bacterial Strains, Plasmids, and Media. Pst strain DC3000and Pseudomonas syringae pv. glycinea (Psg) race 5 strain werekindly provided by D. Cuppels (Agriculture Canada, LondonResearch Centre, London, ON Canada) and N. Keen (Uni-versity of California, Riverside), respectively. Escherichia colistrain DH5a (Bethesda Research Laboratories) was used forcloning procedures. Pseudomonas and E. coli strains werecultured as described (15). Antibiotics (Sigma) were used forselection at the following concentrations: rifampicin, 100,ug/ml; kanamycin, 25 j,g/ml; tetracycline, 15 ,ug/ml; cyclo-heximide, 50 ,ug/ml.

Plasmid pV288 (16) carried avrRpt2 (15, 17), pVB01 (thisstudy) carried avrB (18), pVARM (this study) carried avrRpml(19), and pPPY424 carried avrPph3 (20). All plasmids wereconstructed from the broad-host-range, stable vector pVSP61(W. Tucker, DNA Plant Technology, Oakland, CA), exceptpPPY424, which was constructed from pLAFR3 (18). PlasmidpVB01 was constructed by inserting a 1.3-kb BamHI-Bgl II

fragment containing the entire avrB gene from pPgO-13 (D.Dahlbeck, personal communication) into the BamHI site ofpVSP61. pVARM was constructed by ligating a 2.2-kb EcoRI-Hindlll fragment harboring the entire avrRpml gene frompRPM1 (19) into the EcoRI and HindIll sites of pVSP61. Allconstructs were conjugated into Pseudomonas strains by tri-parental matings using the helper plasmid pRK2013 (21).Growth of Plants, Bacterial Inoculations, and Bacterial

Growth Curves.A. thaliana accessions were obtained from the

Abbreviations: Pst, Pseudomonas syringae pv. tomato; Psg, Pseudomonassyringae pv. glycinea; HR, hypersensitive response; PR, pathogenesis-related; SAR, systemic acquired resistance; CAPS, cleaved amplifiedpolymorphic sequence; RFLP, restriction fragment length polymorphism;cfu, colony-forming unit(s); cM, centimorgan(s).§To whom reprint requests should be addressed.

6597

Dow

nloa

ded

by g

uest

on

July

23,

202

1

6598 Plant Biology: Century et al.

Arabidopsis Biological Resource Center (Ohio State Univer-sity, Columbus). Plants were grown for inoculation as de-scribed (15, 16). For pipet infiltration to assay for disease,plastic transfer pipets were used to introduce bacterial sus-pensions of _ 106 colony-forming units (cfu)/ml in 10 mMMgCl2 into the mesophyll of individual intact leaves. Diseasesymptoms were scored 5 days after inoculation. For the HRassay, bacterial suspensions of -2 x 107 cfu/ml for Pst strainsand 108 cfu/ml for Psg strains were pipet-infiltrated andleaves were scored for tissue collapse 24-36 hr later. Fortesting susceptibility of large numbers of plants, an immersioninoculation procedure was used (16, 22). Disease symptoms ofwatersoaked lesions surrounded by chlorosis were scored 4-5days after inoculation. To determine levels of bacterial growthin planta, leaves of at least six plants per strain were vacuum-infiltrated with bacterial suspensions of i105 cfu/ml. Bacterialgrowth was monitored as described by Whalen et al. (22).

Screen for Susceptible Mutants. Seeds of the accessionColumbia (Col-0) were mutagenized by fast-neutron bombard-ment (courtesy of J. Dangl, University of North Carolina,Chapel Hill). Seeds from 200-300 Ml plants were pooled intosets. To identify mutants in disease resistance response path-ways, we screened 1500 M2 seeds from one larger set of 15pooled sets with DC3000(avrB) by immersion inoculation. M3progeny of putative mutants were retested for susceptibility bypipet infiltration and immersion inoculation.

Fungal Inoculations. The cotyledon assay used for inocu-lations with P. parasitica was described previously (14, 23). AllP. parasitica isolates were derived from oospores obtainedfrom wild populations of A. thaliana. They were derived ineach case from infection of an individual seedling by what wasassumed 'to be a single oospore. The isolates Cala2, Emoy2,Emwal, Hiksl, and Noksl were described previously (14).CandS was derived from the same oospore inoculum as Cala2,and Maks9 was derived from oospore inoculum collected froma household garden in Maidstone, Kent, U.K. Strain P-006 wasdescribed previously (24).

Genetic Analysis. Crosses were performed by hand-emasculating flowers before anther dehiscence and brushingdonor pollen over the carpels. F2 progeny were allowed to

self-fertilize and the seeds were harvested from individual F2plants forming F3 families used for genetic mapping. Geneticmapping was performed with F2 and F3 progeny from anndrl-l x Ler-O cross. Initial genetic mapping was performedwith codominant cleaved amplified polymorphic sequence(CAPS) markers (25). Restriction fragment length polymor-phism (RFLP) markers (26) were obtained through the Ara-bidopsis Biological Resource Center (Ohio State University,Columbus). Restriction enzymes (New England Biolabs) wereused according to the manufacturer's instructions. Standardmethods were used for plasmid/cosmid extractions, prepara-tion of radiolabeled probes by random priming, and DNA gelblot analysis (27, 28). Plant DNA was isolated from leaf tissueof individual F2 plants for minipreparations and root tissue ofseedlings that had been grown in liquid Gamborg's B-5 me-dium (29) for large-scale preparations of F3 family DNA.Multipoint linkage analysis was performed with MAPMAKERMACINTOSH (version 1.0; ref. 30). Map distances in centimor-gans (cM) were calculated from recombination frequencies bythe Kosambi function (31).

RESULTSIdentification of a Mutant Altered in Resistance to Pst. The

A. thaliana accession Col-0 is susceptible to Pst strain DC3000but resistant to DC3000 expressing any one of the avirulencegenes avrB, avrRpml, avrRpt2, and avrPph3. Approximately1500 M2 plants from fast-neutron-mutagenized seeds werescreened for disease susceptibility by immersion inoculationwith DC3000(avrB). Analysis of the self-progeny of one pu-tative mutant demonstrated that its susceptible phenotype washeritable. M3 progeny of the mutant also were inoculated withDC3000 carrying avrRpml, avrRpt2, avrPph3, or the emptyvector (pVSP61) by pipet infiltration and immersion inocula-tion. Four to 5 days after inoculation, the mutant plantsdeveloped watersoaked lesions surrounded by chlorosis inresponse to all of the strains (Fig. 1A). Wild-type Col-0 plantswere susceptible only to DC3000(pVSP61). The mutant ap-peared to harbor a lesion in a locus required for a commonresistance pathway associated with four different avirulence

FIG. 1. Response of ndrl-l and wild-type Col-0 plants to immersion inoculation with DC3000 carrying cloned avirulence genes. Five-week-oldplants were immersion-inoculated with the indicated strains and photographed 5 days later. (A) Leaf of ndrl-l plant inoculated withDC3000(avrRpt2), displaying typical disease symptoms of watersoaked lesions surrounded by chlorosis. (B) Leaf of ndrl-l plant inoculated withDC3000(avrRpm1), showing necrotic patches. (C) Leaf of wild-type Col-0 plant inoculated with DC3000(avrRpml), showing no visible reaction.

Proc. Natl. Acad. Sci. USA 92 (1995)

Dow

nloa

ded

by g

uest

on

July

23,

202

1

Proc. Natl. Acad. Sci. USA 92 (1995) 6599

genes. We have designated the locus NDRJ, for non-race-specific disease resistance, and the mutant allele as ndrl-1.

Interestingly, immersion-inoculated ndrl-J plants exhibitedatypical symptoms in addition to disease lesions in response toDC3000(avrRpml), DC3000(avrB), and DC3000(avrPph3).Outer, older leaves of ndrl-1 plants inoculated with these threestrains developed large necrotic patches, and inner leavesdeveloped small, dry, necrotic spots (Fig. 1B). The necroticpatches were largest and most numerous with DC3000-(avrRpml). The watersoaked lesions caused by these strains onndrl-l plants were fewer in number than those caused byDC3000(pVSP61). Wild-type Col-0 plants did not exhibit thenecrotic symptoms (Fig. 1C), nor did ndrl-1 plants challengedwith DC3000(avrRpt2) or DC3000(pVSP61).

Bacterial Growth in Planta. One crucial characteristic ofresistant plants is the ability to restrict the in planta growth ofavirulent bacteria. To quantify the effect of the ndrl-l muta-tion on bacterial growth, we monitored the growth ofDC3000(avrB), DC3000(avrRpml), and DC3000(avrRpt2) inleaves of intact plants and compared it with the growth ofDC3000 harboring the empty vector, pVSP61 (Fig. 2). Inwild-type Col-0 plants 4 days after inoculation, the populationof DC3000(pVSP61) was 15- to 100-fold greater than that ofDC3000 harboring any one of the three cloned avirulencegenes tested. The ndrl-l plants, however, supported suscepti-ble levels of DC3000 regardless of whether or not a clonedavirulence gene was present. Interestingly, in approximatelyhalf of the independent experiments, DC3000 grew to slightlyhigher levels in ndrl-l plants than in wild-type Col-0 plants,demonstrating a possible increased susceptibility to DC3000 inthe ndrl-l plants.

Induction of a HR-Like Response in ndrl-l Plants. Inresistant A. thaliana accessions, a visible HR usually occurs18-24 hr after leaves are pipet-infiltrated with a suspension ofavirulent bacteria at a concentration of 107 cfu/ml or higher(15). To determine whether the ndrl-l mutation, which ren-

dered the plant susceptible as shown by disease assays, alsoabolished the HR, leaves of wild-type and ndrl-l plants werepipet-infiltrated with bacterial suspensions of -2 x 107 cfu/mland scored for tissue collapse 1 day after infiltration (Table 1).Surprisingly, the ndrl-l mutation appeared to alter the plant'sHR only to DC3000(avrRpt2). DC3000 expressing avrRpml,

9

,S 8,

-7

.C

av

4

20 1 2 3 4 0 1 2 3 4

Days post-inoculation

FIG. 2. Growth of Pst strains in wild-type Col-0 (filled symbols) andndrl-J plants (open symbols). DC3000 (_105 cfu/ml) expressing theavirulence genes avrB (*, O), avrRpmJ (V, 7), and avrRpt2 (-, o) orno cloned avirulence gene (0, 0) was vacuum-infiltrated into plants.Concentrations of bacteria in leaves were assayed 0, 2, and 4 days afterinoculation. DC3000 not expressing a cloned avirulence gene carriedthe empty vector, pVSP61. Data represent the mean SD of triplicatesamples. DC3000(avrB) and DC3000(avrRpml) (A) were assayed inseparate experiments from DC3000(avrRpt2) (B); results cannot bedirectly compared. Experiments were done at least three times for eachavirulence gene tested, with similar results.

Table 1. Response of Col-0 wild-type and ndrl-J plants to the HRassay with Pst strains

HR response

Col-0 line avrB avrRpml avrRpt2 avrPph3 pVSP61

Wild type + + + +ndrl-l + + - +

Pst strain DC3000 carrying the indicated avirulence genes waspipet-infiltrated into individual leaves at 2 x 107 cfu/ml. DC3000 notexpressing a cloned avirulence gene carried the empty vector, pVSP61.HR phenotypes were scored 24 hr after infiltration: +, necrotic tissuecollapse; -, no visible loss of tissue integrity.

avrB, or avrPph3 caused a HR-like response on both wild-typeCol-0 and ndrl-l plants 24 hr after infiltration.Aside from the timing of the response, a HR and a suscep-

tible reaction can look very similar by pipet infiltration on A.thaliana. We therefore suspected that the tissue collapse wasdue to enhanced susceptibility of the mutant to the specificstrains, and not an actual HR. To test this possibility, a Psg race5 strain was used for similar inoculations to assess the HR. Psgcauses neither disease symptoms nor a HR on A. thaliana.However, Psg expressing a cloned avirulence gene that isrecognized by A. thaliana will generally cause a HR onaccessions with the corresponding resistance gene (32). Sincethere is no possibility for interference by early disease symp-toms, inoculum concentrations can be higher and the reactionscan be scored at a later time point, allowing for furtherdevelopment and clarity of the HR. When pipet-infiltrated at108 cfu/ml, Psg carrying avrRpml induced a HR-like re-

sponse in both wild-type Col-0 and ndrl-l plants 36 hr afterinfiltration, while Psg(avrRpt2) induced a HR only in wild-typeCol-0 (data not shown). Psg(pVSP61) caused no reaction ineither plant. Both Psg(avrB) and Psg(avrPph3) induced aHR-like response in ndrl-l plants similar to the responseelicited by inoculation with Psg(avrRpml). ndrl-l plants didnot appear to be susceptible to any of the Psg strains, as shownby immersion inoculations.The ndrl-1 Mutation Alters Resistance to P. parasitica

Isolates. To ascertain whether the ndrl-l mutation altered theplant's resistance to pathogens other than bacteria, the mutantwas tested for response to isolates of the fungal pathogen P.parasitica. The interaction between P. parasitica and A. thali-ana has recently become an important model for investigatingthe molecular basis and evolution of genotype specificity ofplant defense (14, 33-35).

In this study, cotyledons ofndrl-l and wild-type Col-0 plantswere inoculated with eight isolates of P. parasitica and com-pared quantitatively for asexual sporulation based on thenumber of sporangiophores produced on host tissue. Theisolates were either virulent (Maks9 and Noksl) or avirulent(Cala2, Emoy2, Emwal, Cand5, and Hiksl) on wild-typeCol-0. A statistically significant increase in sporulation by eachof the avirulent isolates was observed on ndrl-l compared withwild-type (Table 2). An avirulent isolate obtained from Bras-sica oleracea (P-006) was also included in the experiment andexhibited no sporulation on either ndrl-l or wild-type Col-0plants.

Genetic Analysis. To determine the genetic basis for sus-ceptibility in the mutant line, ndrl-l plants were crossed to thefollowing lines: Col-0 gll, a morphologically marked line thatlacks trichomes (36); Bla-2, an accession with a recessive alleleat the RPS3/RPMI locus rendering it susceptible to DC3000-(avrB) and DC3000(avrRpml) (32); and Ler-0, an accessionresistant to DC3000(avrB), DC3000(avrRpml), and DC3000-(avrRpt2). F1 plants were assayed for disease resistance bypipet infiltration and F2 plants were evaluated by immersioninoculation. F1 plants of all crosses were fully resistant toDC3000 carrying any one of the avirulence genes tested,indicating that the parent lines crossed to the mutant carried

A j B

Plant Biology: Century et al.

Dow

nloa

ded

by g

uest

on

July

23,

202

1

6600 Plant Biology: Century et al.

Table 2. Asexual reproduction (measured as number of sporangiophores) by eight isolates of P.parasitica 7 days after inoculation of cotyledons of wild-type Col-0 and ndrl-l

No. of sporangiophores per cotyledon,

P. parasitica mean ± SE (n)t t-test statisticstisolate RPP locus* Wild-type ndrl-l t value df P

Cala2 2-IV 0.2 ± 0.13 (45) 1.5 ± 0.49 (39) 2.61 82 0.01Emoy2 4-IV 1.4 ± 0.20 (55) 18.0 ± 0.60 (44) 28.76 97 0.00Emwal 4-IV 1.4 ± 0.19 (50) 18.9 ± 0.28 (45) 53.11 93 0.00CandS 7-I 2.5 ± 0.38 (44) >20.0 (27) 37.12 70 0.00Hiksl 7-I 0.0 ± 0.00 (45) 0.3 ± 0.11 (35) 3.35 78 0.00Maks9 None 18.8 ± 0.58 (38) 19.3 ± 0.51 (38) 0.58 74 0.57Noksl None >20.0 (50) >20.0 (49) 0.00 97 1.00P-006 Unknown 0.0 ± 0.00 (25) 0.0 ± 0.00 (25) 0.00 48 1.00

*Loci associated with recognition of P. parasitica, identified here by the locus number and chromosome(Roman numeral). See ref. 14 for further details.tMaximum of 20 cotyledons were counted; n, no. of seedlings inoculated.lThe t value is pooled; df, degrees of freedom; P, probability value.

functional wild-type alleles ofNDRI and that the ndrl-l allelewas recessive to the wild-type allele(s). The F1 progeny fromthe Bla-2 x ndrl-l cross were resistant to DC3000(avrB) andDC3000(avrRpml), demonstrating the lack of alterations atthe RPS3/RPMI locus of the ndrl-l parent. F2 progeny fromthis cross were pipet-inoculated with DC3000 expressing avrB,avrRpml, or avrRpt2. In all cases, F2 plants which weresusceptible to DC3000(avrRpt2) were also susceptible toDC3000(avrB) or DC3000(avrRpml), indicating that a singlelocus was responsible for the phenotype. Analysis of F2progeny from crosses of ndrl-l to Col-0 gll and Ler-O dem-onstrated that the general susceptible phenotype was inheritedas a single Mendelian locus. Resistance to DC3000(avrB)segregated -3 resistant:1 susceptible in the F2 progeny of boththe Col-Ogll x ndrl-l (156:63; Xy = 1.66; P > 0.1) and ndrl-lx Ler-O (579:186; x2 = 0.192; P > 0.5) crosses.Progeny from the ndrl-l X Ler-O cross were used to

genetically map ndrl-l, because of the large number of mark-ers available which detect polymorphisms in Col-O and Ler-0.Initial mapping was done by using the CAPS mapping tech-nique (25) on 54 susceptible F2 plants. ndrl-l was located onchromosome III, linked to CAPS marker gll by =18 cM.Seventy-seven F3 families that had been scored for theirresistance phenotype were then used to map ndrl-l relative toRFLP markers in the area surrounding gll. ndrl-l was foundto be -2.0 cM from marker g6220 and -6.5 cM from markerg4711.

DISCUSSIONWe have used a genetic approach to dissect the molecularpathways leading to disease resistance in A. thaliana. We haveidentified a mutant (ndrl-l) of the accession Col-0 which isaltered in resistance to both a bacterial and a fungal pathogenin a non-race-specific manner. The ndrl-l mutation suggeststhat a common step exists in pathways of resistance not onlyto different isolates within a pathogen species but also tospecies from entirely different phylogenetic kingdoms, includ-ing both a eukaryote and a prokaryote. Resistance to thepathogens Psg and P. parasitica race P-006, for which A.thaliana is not a host, is not altered, suggesting that NDRJ isnot required for non-host resistance to these pathogens.There are several possible ways thatNDR1 might be involved

in disease resistance. We observed by RNA blot analysis thataccumulation of RPS2 transcripts was not altered in ndrl-lplants compared with wild type (data not shown), suggestingthat NDR1 is not required for expression of race-specificresistance genes. We propose that NDRI is required in acommon signal transduction pathway downstream of the initialrecognition of an avirulent pathogen. It is possible that theputative gene product of NDRJ interacts directly with RPS2

and/or other resistance gene products in the same family, orwith other components in a signal transduction pathwaydownstream of the race-specific resistance gene products.Additionally, NDRJ may be involved in disease resistancepathways requiring salicylic acid. Salicylic acid has been shownto be an important component in systemic acquired resistance(SAR) in plants (37, 38). Delaney et al. (39) have suggestedthat salicylate is also required in A. thaliana for race-specificdisease resistance. Transgenic Col-0 plants producing salicy-late hydroxylase, which degrades salicylic acid, are phenotyp-ically similar to ndrl-l plants with regard to enhanced suscep-tibility to Pst and P. parasitica strains. However, virulent Pststrains grow to higher levels than an avirulent strain insalicylate hydroxylase-expressing plants, while this is not ob-served in ndrl-l plants.When pipet-infiltrated at 2 x 107 cfu/ml, DC3000 carrying

avrRpml, avrB, or avrPph3 causes a HR-like collapse of tissuein ndrl-l plants. The necrotic patches caused by these samestrains after immersion inoculation may be a less severemanifestation of the HR-like response observed in the HRassay. The HR-like phenotype observed on ndrl-l plants isintriguing because it suggests that a HR can be induced withouttriggering the steps leading to limitation of pathogen growth.Interestingly, the acd2 (40) and Isd (41) mutants ofA. thalianadevelop spontaneous, wound-induced, or pathogen-inducedHR-like lesions similar to the necrotic patches induced byimmersion inoculation of ndrl-l plants. Unlike ndrl-l plants,however, acd2 and Isd mutants are less susceptible thanwild-type plants to Pseudomonas syringae pathovars.The avirulence-gene specificity of the HR-like response in

ndrl-l plants possibly can be explained by functional differ-ences in the resistance genes themselves. Different avirulencegenes induce qualitatively different HRs on variousA. thalianaaccessions (unpublished observation). It is possible that a givenspecific recognition event initiates numerous secondary eventswhich condition the overall resistance phenotype. One or moreof these secondary events may require NDR1, and differentresistance gene products may differentially induce the varioussecondary pathways. In the case of avrRpt2, the induction ofthe HR absolutely requires NDR1, while with avrB, avrRpml,or avrPph3 putative alternative pathways not requiring NDR1still trigger the HR.

Similar to what we find, others have demonstrated theuncoupling of plant defense responses from each other. Ja-kobek and Lindgren (42) demonstrated that bean plantsinoculated with a Pseudomonas syringae pv. tabaci hrp mutant(which no longer causes a HR) or nonpathogenic bacteriarapidly accumulate transcripts of genes in the phenylpropanoidpathway even though a visible HR is not induced. Normally theHR is associated with the rapid activation of these genes. Caoet al. (43) recently reported the identification of anA. thaliana

Proc. Natl. Acad. Sci. USA 92 (1995)

Dow

nloa

ded

by g

uest

on

July

23,

202

1

Proc. Natl. Acad. Sci. USA 92 (1995) 6601

mutant, nprl, which does not express PR genes or developSAR after treatment with salicylic acid or its analog 2,6-dichloroisonicotinic acid. Pseudomonas syringae pv. maculicolabacteria expressing avrRpt2 inoculated on nprl plants induce aHR, although SAR is not exhibited. Usually, SAR is correlatedboth with a HR and with PR gene induction (37). The padmutants ofA. thaliana (44), which are deficient in productionof the phytoalexin camalexin, are not altered in resistance toavirulent Pseudomonas syringae strains based on in plantabacterial growth, but they do allow more growth of the virulentstrain. Clearly, a big question remains as to which defense-related responses are necessary and sufficient for limitation ofpathogen growth. That these different defense-related re-sponses can be separated from one another suggests thatmultiple signal transduction pathways are used. Our work onndrl-l indicates that some components may be conserved inthese pathways. Further analysis of the ndrl-l mutation,identification of other alleles and loci, and cloning of theNDRIgene should provide a better understanding of the componentsrequired for disease resistance and elucidate some of the stepsinvolved in the resistance process.

We thank Jeff Dangl for providing mutagenized seed; John Lucasfor providing the isolate P-006; and Maureen Whalen, Tom Tai, andAllan Shapiro for critical reading of the manuscript. This work wassupported by the National Science Foundation (a predoctoral fellow-ship to K.S.C. and Grant MCB-9219959 to B.J.S.). The Peronosporawork was supported financially by the United Kingdom Agricultureand Food Research Council.

1. Flor, H. (1971) Annu. Rev. Phytopathol. 9, 275-296.2. Keen, N. T. (1990) Annu. Rev. Genet. 24, 447-463.3. Klement, Z. (1982) in Hypersensitivity, eds. Mount, M. S. & Lacy,

G. H. (Academic, New York), pp. 149-177.4. Lamb, C. J., Lawton, M. A., Dron, M. & Dixon, R. A. (1989) Cell

56, 215-224.5. Dixon, R. A. & Lamb, C. J. (1990)Annu. Rev. Plant Physiol. Plant

Mol. Biol. 41, 339-367.6. Keen, N. & Staskawicz, B. (1988) Annu. Rev. Microbiol. 42,

421-440.7. Martin, G. B., Brommonschenkel, S. H., Chunwongse, J., Frary,

A., Ganal, M. W., Spivey, R., Wu, T., Earle, E. D. & Tanksley,S. D. (1993) Science 262, 1432-1436.

8. Bent, A. F., Kunkel, B. N., Dahlbeck, D., Brown, K. L., Schmidt,R., Giraudat, J., Leung, J. & Staskawicz, B. (1994) Science 265,1856-1860.

9. Mindrinos, M., Katagiri, F., Yu, G.-L. & Ausubel, F. M. (1994)Cell 78, 1089-1099.

10. Whitham, S., Dinesh-Kumar, S. P., Choi, D., Hehl, R., Corr, C.& Baker, B. (1994) Cell 78, 1101-1115.

11. Jones, D. A., Thomas, C. M., Hammond-Kossack, K. E., Balint-Kurti, P. J. & Jones, J. D. G. (1994) Science 266, 789-793.

12. Kobe, B. & Deisenhofer, J. (1994) Trends Biochem. Sci. 19,415-421.

13. Dangl, J. L. (1993) in The Emergence ofArabidopsis thaliana as aModel for Plant-Pathogen Interactions, eds. Andrews, J. H. &Tommerrup, I. (Academic, London), Vol. 10, pp. 127-156.

14. Holub, E. B., Beynon, J. L. & Crute, I. R. (1994) Mol. PlantMicrobe Interact. 7, 223-239.

15. Whalen, M., Innes, R., Bent, A. & Staskawicz, B. (1991) PlantCell 3, 49-59.

16. Kunkel, B. N., Bent, A. F., Dahlbeck, D., Innes, R. W. &Staskawicz, B. J. (1993) Plant Cell 5, 865-875.

17. Dong, X., Mindrinos, M., Davis, K. R. & Ausubel, F. M. (1991)Plant Cell 3, 61-72.

18. Staskawicz, B., Dahlbeck, D., Keen, N. & Napoli, C. (1987) J.Bacteriol. 169, 5789-5794.

19. Debener, T., Lehnackers, H., Arnold, M. & Dangl, J. (1991) PlantJ. 1, 289-302.

20. Jenner, C., Hitchin, E., Mansfield, J., Walters, K., Betteridge, P.,Teverson, D. & Taylor, J. (1991) Mo. Plant Microbe Interact. 4,553-562.

21. Figurski, D. & Helinski, D. R. (1979) Proc. Natl. Acad. Sci. USA76, 1648-1652.

22. Whalen, M. C., Stall, R. E. & Staskawicz, B. J. (1988) Proc. Natl.Acad. Sci. USA 85, 6743-6747.

23. Dangl, J. L., Holub, E. B., Debener, T., Lehnackers, H., Ritter,C. & Crute, I. R. (1992) in Genetic Definition ofLoci Involved inArabidopsis-Pathogen Interactions, eds. Koncz, C., Chua, N. &Schell, J. (World Scientific, Singapore), pp. 393-418.

24. Moss, N. A., Crute, I. R. & Lucas, J. A. (1994) Plant Pathol. 43,713-725.

25. Konieczny, A. & Ausubel, F. M. (1993) Plant J. 4, 403-410.26. Nam, H.-G., Giraudat, J., Den Boer, B., Moonan, F., Loos,

W. D. B., Hauge, B. M. & Goodman, H. M. (1989) Plant Cell 1,699-706.

27. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989) MolecularCloning: A Laboratory Manual (Cold Spring Harbor Lab. Press,Plainview, NY), 2nd Ed.

28. Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Sei-dman, J. G., Smith, J. A. & Struhl, K., eds. (1987) CurrentProtocols in Molecular Biology (Greene and Wiley, New York).

29. Gamborg, 0. L., Miller, R. A. & Ojima, K. (1968) Exp. Cell Res.50, 151-158.

30. Lander, E. S., Green, P., Abrahamson, J., Barlow, A., Daly, M. J.,Lincoln, S. E. & Newburg, L. (1987) Genomics 1, 174-181.

31. Kosambi, D. D. (1944) Ann. Eugen. 12, 172-175.32. Innes, R. W., Bisgrove, S. R., Smith, N. M., Bent, A. F., Staskaw-

icz, B. J. & Liu, Y. C. (1993) Plant J. 4, 813-820.33. Koch, E. & Slusarenko, A. (1990) Plant Cell 2, 437-445.34. Parker, J. E., Szabo, V., Staskawicz, B. J., Lister, C., Dean, C.,

Daniels, M. J. & Jones, J. D. G. (1993) Plant J. 4, 821-831.35. Tor, M., Holub, E. B., Brose, E., Musker, R., Gunn, N., Can, C.,

Crute, I. R. & Beynon, J. L. (1994) Mol. Plant Microbe Interact.7, 214-222.

36. Koornneef, M., Dellaert, L. W. M. & van der Veen, J. H. (1982)Mutat. Res. 93, 109-123.

37. Malamy, J. & Klessig, D. F. (1992) Plant J. 2, 643-654.38. Gaffney, T., Friedrich, L., Vernooij, B., Negrotto, D., Nye, G.,

Uknes, S., Ward, E., Kessmann, H. & Ryals, J. (1993) Science 261,754-756.

39. Delaney, T. P., Uknes, S., Vernooij, B., Friedrich, L., Weymann,K., Negrotto, D., Gaffney, T., Gut-Rella, M., Kessmann, H.,Ward, E. & Ryals, J. (1994) Science 266, 1247-1250.

40. Greenberg, J. T., Guo, A., Klessig, D. F. & Ausubel, F. M. (1994)Cell 77, 551-563.

41. Dietrich, R. A., Delaney, T. P., Uknes, S. J., Ward, E. R., Ryals,J. A. & Dangl, J. L. (1994) Cell 77, 566-577.

42. Jakobek, J. L. & Lindgren, P. B. (1993) Plant Cell 5, 49-56.43. Cao, H., Bowling, S. A., Gordon, A. S. & Dong, X. (1994) Plant

Cell 6, 1583-1592.44. Glazebrook, J. & Ausubel, F. M. (1994) Proc. Natl. Acad. Sci.

USA 91, 8955-8959.

Plant Biology: Century et al.

Dow

nloa

ded

by g

uest

on

July

23,

202

1