Identification of Three Putative Signal Transduction Genes Involved in R Gene-Specified Disease Resistance in Arabidopsis

Corresponding author: Roger W. Innes, Department of Biology, Indiana University, Bloomington, IN 47405., rinnes{at}bio.indiana.edu (E-mail)

Communicating editor: J. CHORY

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The RPS5 disease resistance gene of Arabidopsis mediates recognition of Pseudomonas syringae strains that possess the avirulence gene avrPphB. By screening for loss of RPS5-specified resistance, we identified five pbs (avrPphB susceptible) mutants that represent three different genes. Mutations in PBS1 completely blocked RPS5-mediated resistance, but had little to no effect on resistance specified by other disease resistance genes, suggesting that PBS1 facilitates recognition of the avrPphB protein. The pbs2 mutation dramatically reduced resistance mediated by the RPS5 and RPM1 resistance genes, but had no detectable effect on resistance mediated by RPS4 and had an intermediate effect on RPS2-mediated resistance. The pbs2 mutation also had varying effects on resistance mediated by seven different RPP (recognition of Peronospora parasitica) genes. These data indicate that the PBS2 protein functions in a pathway that is important only to a subset of disease-resistance genes. The pbs3 mutation partially suppressed all four P. syringae-resistance genes (RPS5, RPM1, RPS2, and RPS4), and it had weak-to-intermediate effects on the RPP genes. In addition, the pbs3 mutant allowed higher bacterial growth in response to a virulent strain of P. syringae, indicating that the PBS3 gene product functions in a pathway involved in restricting the spread of both virulent and avirulent pathogens. The pbs mutations are recessive and have been mapped to chromosomes I (pbs2) and V (pbs1 and pbs3).

PATHOGEN resistance in plants is often characterized by a gene-for-gene relationship requiring a specific resistance (R) gene from the plant and a corresponding avirulence (avr) gene from the pathogen (FLOR 1971 ). The presence of the appropriate R-avr gene pair results in host resistance, whereas the absence or inactivation of either member of the gene pair results in susceptibility of the host to the pathogen. Although the molecular mechanisms are still unknown, R genes mediate specific recognition of pathogens, presumably by some sort of receptor-ligand interaction (GABRIEL and ROLFE 1990 ).

After initial pathogen recognition, signaling events that result in the activation of plant defenses and the limitation of pathogen growth are triggered. These defense responses are often correlated with rapid, localized necrosis at the site of infection (hypersensitive response), and they include an oxidative burst, cell wall fortification, production of antimicrobial compounds (phytoalexins), and the accumulation of pathogenesis-related proteins (HAMMOND-KOSACK and JONES 1996 ). R genes that mediate resistance to bacterial, fungal, oomycete, viral, and nematode pathogens have been cloned from several plant species (reviewed in BENT 1996 ; BAKER et al. 1997 ; JONES and JONES 1997 ; ELLIS and JONES 1998 ). Structural motifs are shared among many of these R gene proteins, indicating that disease resistance to diverse pathogens may operate through similar molecular pathways. However, components of these pathways and their function remain largely undefined.

To identify potential signal transduction components used by R genes, we and others have screened mutagenized plants for loss of resistance to specific pathogens (see reviews by HAMMOND-KOSACK and JONES 1996 ; KUNKEL 1996 ; BAKER et al. 1997 ; INNES 1998 ). By design, these screens also identify mutations within R genes. In fact, the isolation of R gene mutants appears to be much more common than the identification of mutants involved in R gene-mediated signal transduction pathways (INNES 1998 ). This may indicate that some pathway components are redundant or required for viability. Additionally, these pathways may be branched such that a particular mutation abolishes only a subset of defense responses. Such mutants may have been overlooked because pathogen resistance was not completely compromised. To date, only a small number of putative R gene signal transduction mutants have been identified from genetic screens for loss of pathogen resistance. These include mutants in barley (rar1 and rar2; TORP and JORGENSEN 1986 ; JORGENSEN 1988 ; FREIALDENHOVEN et al. 1994 ), tomato (rcr1, rcr2, and prf); (HAMMOND-KOSACK et al. 1994 ; SALMERON et al. 1994 ), and Arabidopsis (ndr1 and eds1); (CENTURY et al. 1995 ; PARKER et al. 1996 ).

Most of these mutations affect the function of a subset of R genes tested. The rar1 and rar2 mutations reduce resistance conferred by several powdery mildew-resistance genes (FREIALDENHOVEN et al. 1994 ; JORGENSEN 1996 ). The ndr1 and eds1 mutations affect resistance against both bacterial and oomycete pathogens (CENTURY et al. 1995 , CENTURY et al. 1997 ; PARKER et al. 1996 ; AARTS et al. 1998 ). Interestingly, R genes that are strongly suppressed by one of these Arabidopsis mutations are not greatly affected by the other mutation, indicating that NDR1 and EDS1 may be critical for different signaling pathways (AARTS et al. 1998 ). The rcr1, rcr2, and prf mutations are reported to affect pathogen resistance conferred by single R genes, but their effect on the function of other R genes has not been tested extensively (HAMMOND-KOSACK et al. 1994 ; SALMERON et al. 1994 ).

Although the identified mutations reduce resistance conferred by specific R genes, most do not eliminate all defense responses. In rar1 and rar2 plants, this is demonstrated as an intermediate level of susceptibility to powdery mildew (TORP and JORGENSEN 1986 ; JORGENSEN 1988 , JORGENSEN 1996 ). The rcr1 and rcr2 mutations weaken resistance against Cladosporium fulvum, but neither mutation allows sporulation of the fungus (HAMMOND-KOSACK et al. 1994 ). Plants carrying the ndr1 mutation allow extensive growth of several previously avirulent races of Pseudomonas syringae pv tomato; however, these plants can still produce a visible hypersensitive response to some of these bacterial strains (CENTURY et al. 1995 ). Additionally, ndr1 plants are only partially susceptible to most isolates of Peronospora parasitica (CENTURY et al. 1995 , CENTURY et al. 1997 ).

To identify additional components of R gene signal transduction pathways, we screened for mutations that suppressed resistance mediated by the RPS5 gene of Arabidopsis. RPS5 confers resistance to P. s. tomato carrying avrPphB (formerly called avrPph3, SIMONICH and INNES 1995 ). In contrast to previous screens performed with avrRpt2 and avrB (KUNKEL et al. 1993 ; BISGROVE et al. 1994 ), we recovered mutations in three loci other than the targeted R gene. Here we describe the effect of these mutations on RPS5 and on several other R genes that confer resistance to various strains of P. s. tomato and P. parasitica.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Pseudomonas strains and Peronospora isolates:
P. syringae strains were cultured as described previously (INNES et al. 1993 ). P. s. tomato strains carrying avrB, avrB::, avrRpt2, avrRps4, and avrPphB have been described (INNES et al. 1993 ; SIMONICH and INNES 1995 ; HINSCH and STASKAWICZ 1996 ). The P. parasitica isolates and their cultivation have also been described previously (DANGL et al. 1992 ; HOLUB et al. 1994 ).

Growth of plants, plant inoculations, and bacterial growth curves:
Growth conditions for Arabidopsis were as described previously (BISGROVE et al. 1994 ). Mutagenized seeds (M₂ generation) were obtained from M. Estelle (ethyl methanesulfonate-mutagenized and gamma-irradiated seeds), and Lehle seeds (Round Rock, TX; fast-neutron-mutagenized seeds). In all cases, mutagenesis was performed on seeds (M₁ generation), and the plants were allowed to self-fertilize. Seeds from ~500 M₁ plants were pooled to generate bulked M₂ seed lots. A total of 32 lots were screened to identify the pbs mutants. The pbs1-1 mutation was induced by fast neutrons, the pbs1-2 and pbs3 mutants were induced by EMS, and the two pbs2 mutants were induced by gamma irradiation. It is assumed that the two pbs2 mutants carry identical mutations because they were isolated from the same seed lot (856 plants were screened from this lot).

Plants were inoculated by dipping whole rosettes in a suspension of ~2 x 10⁸ colony-forming units of P. s. tomato per milliliter as described previously (INNES et al. 1993 ). Genotypes of putative mutants were confirmed as being Col-0 and not contaminating susceptible genotypes through use of several microsatellite and codominant cleaved amplified polymorphic sequences (CAPS) markers (KONIECZNY and AUSUBEL 1993 ; BELL and ECKER 1994 ). To monitor bacterial growth in Arabidopsis leaves, we inoculated plants by vacuum infiltration of 5 x 10⁵ cfu/ml suspension of P. s. tomato, as described by WHALEN et al. 1991 . The surfactant Silwet L-77 (OSi Specialties, Inc., Danbury, CT) was added at a concentration of 0.001%. Samples were removed from rosette leaves, macerated, diluted, and plated on selective medium, as described previously (BISGROVE et al. 1994 ). Colonies were counted 48 hr later.

Resistance of Arabidopsis accessions to P. parasitica was assayed by inoculating seedling cotyledons as described previously (DANGL et al. 1992 ; HOLUB et al. 1994 ). A minimum of 30 seedlings distributed among 5 replications were used per plant genotype/P. parasitica isolate combination in all experiments.

Genetic analysis:
Crosses were performed by hand-emasculating flowers before anther dehiscence and then brushing donor pollen over the stigmas. F₁, F₂, and F₃ plants were scored for disease phenotypes using the dip assay. Seeds were collected from individual selfed F₁ and F₂ plants to generate plants for the next generation. Genetic mapping was performed by polymerase chain reaction using oligonucleotide primers designed to amplify microsatellite sequences (BELL and ECKER 1994 ) or CAPS (KONIECZNY and AUSUBEL 1993 ). These markers have been used by the Arabidopsis community to establish a well-defined genetic map on a set of recombinant, inbred lines derived from a cross between Ler and Col-0 (the Lister-Dean recombinant inbred map, http://genome-www.stanford.edu/Arabidopsis/ww/home.html/). Restriction endonucleases (New England Biolabs, Beverly, MA) were used according to the manufacturer's instructions, and DNA isolation from F₂ leaf tissue was performed as described previously (FRYE and INNES 1998 ). Map distances in centimorgans were calculated from recombination frequencies by the Kosambi function (KOSAMBI 1944 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Isolation of pbs mutants:
To identify disease-resistance mutants, we inoculated ~16,600 mutagenized Col-0 plants by immersion in a suspension of P. s. tomato strain DC3000(avrPphB). Mutants were identified by the presence of disease symptoms 4–5 days after inoculation. We have previously reported the isolation of two mutant plants from this screen that carried mutations within the resistance gene RPS5 (WARREN et al. 1998 ). We also identified five plants that carried mutations in genes other than RPS5. As described below, these five mutants represented three complementation groups, which we have designated pbs1, pbs2, and pbs3 for avrPphB susceptible. Figure 1 shows that pbs1, pbs2, and pbs3 plants developed disease symptoms of chlorosis and water-soaked lesions after infection with DC3000(avrPphB). Wild-type Col-0 plants remained green and healthy. All self-progeny from the mutants were susceptible to DC3000 (avrPphB), indicating that they were homozygous for the mutations.

View larger version (131K): In this window In a new window Download PPT slide   Figure 1. Disease symptoms induced by P. s. tomato strains on pbs mutants. The parental accession, Col-0, and the pbs1-1, pbs2, and pbs3 mutants were infected by brief submersion in DC3000 strains carrying the indicated avirulence genes. refers to strain DC3000(avrB::), which is a virulent control carrying the avrB gene that has been disrupted by the insertion of an fragment. Photographs were taken 5 days after inoculation.

Genetic analysis of the pbs mutants is shown in Table 1. The five mutants were backcrossed to Col-0 plants. All the F₁ plants were resistant to DC3000(avrPphB), indicating that the mutations were recessive. The ratio of resistant to susceptible plants in the F₂ generation was determined for each complementation group (see below). Segregation was consistent with a 3:1 ratio for pbs1, pbs2, and pbs3 plants, indicating that each susceptible phenotype was caused by a single mutation.

To determine that the mutations were not in RPS5, we crossed Col pbs1, Col pbs2, and Col pbs3 plants to the Arabidopsis accession Landsberg erecta (Ler), which naturally lacks RPS5 function (SIMONICH and INNES 1995 ). F₁ plants from these crosses were resistant to DC3000(avrPphB), and plants in the F₂ generation segregated for resistance. These results demonstrated that the pathogen susceptibility exhibited by mutant plants was not caused by a defect present in the RPS5 gene.

The pbs1, pbs2, and pbs3 complementation groups were established by crossing mutant plants to each other. Mutations were considered allelic if all plants from the resulting generations developed disease symptoms in response to DC3000(avrPphB). Of the five mutants isolated, two were placed in the pbs1 complementation group, two were placed in the pbs2 complementation group, and one was placed into the pbs3 complementation group (Table 1). In the case of the two pbs2 mutants, rather than representing different mutant alleles of the same gene, they likely represent the same mutation since they were isolated from the same pool of mutagenized seed (see MATERIALS AND METHODS).

The pbs mutants exhibit decreased resistance to multiple P. s. tomato strains:
In addition to RPS5, Col-0 plants possess the R genes RPS2, RPM1, and RPS4. These R genes confer resistance to P. s. tomato strains carrying avrRpt2, avrB, or avrRps4, respectively (INNES et al. 1993 ; KUNKEL et al. 1993 ; HINSCH and STASKAWICZ 1996 ). To determine if the pbs mutations disrupted the function of these other R genes, we infected Col-0, Col pbs1, Col pbs2, and Col pbs3 plants with DC3000 carrying each of these avr genes.

As shown in Figure 1, Col pbs1-1 plants remained resistant to DC3000 carrying avrRpt2, avrB, or avrRps4. Identical results were obtained with Col pbs1-2 plants (data not shown). We quantified bacterial growth within Col pbs1-1 plants, and these data are shown in Figure 2. Consistent with visible symptoms, Col pbs1-1 plants exhibited enhanced growth only to DC3000(avrPphB). These results are similar to those expected for a mutation in RPS5, and they suggest that PBS1 is part of a signal transduction pathway specific to RPS5.

View larger version (35K): In this window In a new window Download PPT slide   Figure 2. Growth of P. s. tomato strains within leaves of pbs mutants. The parental accession, Col-0, and the pbs1-1, pbs2, and pbs3 mutants were inoculated by vacuum infiltration, with strain DC3000 carrying the indicated avirulence genes. Growth of bacteria within the leaves was monitored over a 4-day time course. Each data point represents the mean ± SE of three samples. Data shown are representative of two independent experiments.

In contrast to Col pbs1 plants, Col pbs2 plants developed disease symptoms after infection with DC3000 carrying avrB or avrRpt2 (Figure 1). Resistance was not fully compromised to DC3000(avrRpt2), which induced less chlorosis and fewer lesions than DC3000(avrPphB) or DC3000(avrB). Col pbs2 plants appeared resistant to DC3000(avrRps4; Figure 1). In separate trials, these plants were either indistinguishable from wild-type Col-0 plants, or they developed mild disease symptoms that could only be scored on a subset of plants. Thus, for resistance controlled by RPS4, the PBS2 gene product probably does not play a significant role. These results were confirmed by bacterial growth curves (Figure 2). In Col pbs2 plants, DC3000(avrPphB) and DC3000(avrB) achieved a level of growth similar to that of a virulent strain of P. s. tomato. DC3000(avrRpt2) showed slightly elevated growth in Col pbs2 plants compared to wild-type plants, whereas growth of DC3000(avrRps4) was similar in both mutant and wild-type plants.

The increased susceptibility of Col pbs2 plants to P. s. tomato strains carrying avrPphB, avrRpt2, or avrB did not appear to be caused by a second site mutation. We infected F₃ families derived from Col pbs2 backcrossed plants with DC3000(avrPphB), DC3000 (avrRpt2), and DC3000(avrB). Ten families obtained from DC3000(avrPphB)-susceptible F₂ plants developed disease symptoms in response to all three bacterial strains, indicating that the phenotypes were caused by the same or closely linked mutations.

Col pbs3 plants developed disease symptoms in response to DC3000 carrying avrRpt2, avrB, or avrRps4 (Figure 1). DC3000(avrRpt2) induced the strongest disease symptoms, and DC3000 carrying avrPphB, avrB, or avrRps4 caused less severe disease symptoms. However, resistance was not fully compromised against any of the avirulent pathogens. The bacterial growth of all four avirulent P. s. tomato strains was elevated in Col pbs3 plants, but did not reach the same level of growth as seen for a virulent strain of P. s. tomato infecting wild-type Col-0 plants (Figure 2). The decreased resistance to all four P. s. tomato strains cosegregated in 15 F₃ families that were derived from either DC3000 (avrPphB)- or DC3000(avrRpt2)-susceptible F₂ plants.

As shown in Figure 3A, unlike Col pbs1 and Col pbs2 plants, Col pbs3 plants developed more severe disease symptoms than wild-type Col-0 plants when infected by DC3000 containing no added avirulence gene. The growth of this virulent P. s. tomato strain in Col pbs3 leaves is quantified in Figure 3B. Bacterial growth was slightly elevated relative to wild-type Col-0 plants in multiple trials and was statistically significant at 2 days after inoculation. These results suggest that the PBS3 gene product may be involved in restricting the growth of both virulent and avirulent pathogens.

View larger version (81K): In this window In a new window Download PPT slide   Figure 3. Response of pbs3 mutant to virulent P. s. tomato. (A) The parental accession, Col-0, and pbs3 were infected by brief submersion in strain DC3000(avrB::), which is a virulent strain of P. s. tomato carrying the avrB gene that has been disrupted by the insertion of an fragment. Photographs were taken 5 days after inoculation. Col-0 and pbs3 plants are shown at the same magnification. (B) Growth of DC3000 (avrB::) within leaves of Col-0 and pbs3 plants was monitored over a 4-day time course after inoculation by vacuum infiltration. Each data point represents the mean ± SE of three samples. Data from two independent experiments are shown.

The pbs mutants exhibit decreased resistance to several P. parasitica isolates:
Because the pbs2 and pbs3 mutations affected resistance to multiple P. s. tomato strains, we tested whether resistance to the biotrophic oomycete P. parasitica (downy mildew) was also affected by the pbs mutations. The degree of resistance was measured by counting the number of sporangiophores (tree-like structures emerging from stomata and bearing conidiosporangia) produced in cotyledons. We assessed sporulation in cotyledons of Col pbs1-2, Col pbs2, and Col pbs3 seedlings by seven isolates of P. parasitica, which are each diagnostic for a different wild-type RPP (recognition of P. parasitica) gene. As shown in Table 2, differences were observed among the three pbs mutants in their response to the seven P. parasitica isolates.

Resistance to each of the isolates appeared to be mostly unaffected in Col pbs1-2 plants. No detectable change from wild type was observed after inoculation with three isolates (Cala2, Hind4, and Hiks1), and a significant but very weak enhanced sporulation was seen with the other four isolates (Emoy2, Wela3, Cand5, and Wand1, Table 2). This enhanced sporulation was much less than that exhibited in a fully susceptible plant, such as Cand5 in Col ndr1, which had a mean of at least 20 sporangiophores per cotyledon.

In contrast, the pbs2 mutation enhanced sporulation to five of the isolates, with an increase to full susceptibility for at least two isolates (Cand5 and Wand1) that produce a rare sporophore or no sporulation, respectively, in wild-type Col-0 cotyledons (Table 2). Medium sporulation was seen with the Emoy2 isolate, and low sporulation was witnessed after inoculation with the Hind4 and Wela3 isolates. Col pbs2 plants appeared very similar to wild type after inoculation with the remaining two isolates, Cala2 and Hiks1.

Col pbs3 plants exhibited a third pattern of altered resistance to the isolates (Table 2). Similar to the results obtained with this mutant after bacterial inoculations, resistance to the P. parasitica isolates was not fully compromised. However, Emoy2 produced a mean of 16 sporophores per cotyledon compared with a mean of 2 in the wild type, and the mutant was significantly altered to a lesser degree in its response to five other isolates. Cala2 was the only isolate that appeared to exhibit no change in phenotype between the mutant and wild type.

In addition to the pbs mutants, we assessed sporulation in Col ndr1 cotyledons. Similar to the pbs mutants, Col ndr1 plants have been previously reported to exhibit decreased resistance to avirulent P. s. tomato strains and downy mildew isolates (CENTURY et al. 1995 , CENTURY et al. 1997 ). Similar to the pbs mutants, the level of asexual reproduction varied from no sporulation to heavy sporulation in Col ndr1 cotyledons, depending on the particular isolate being tested (Table 2). The Col ndr1 mutants, however, exhibited a pattern of responses to the seven P. parasitica isolates that was distinct from that observed in the pbs mutants (Table 2).

The pbs1 and pbs2 mutations map to chromosomes V and I:
Molecular markers were used to determine map positions for the PBS genes. The pbs1-1, pbs1-2, and pbs2 mutations, present in a Col-0 background, were crossed to the accession Ws-0 that possesses RPS5 function. F₂ plants homozygous for pbs1-1, pbs1-2, and pbs2 mutations were selected on the basis of pathogen susceptibility. As in backcrossed plants, susceptibility to DC3000(avrPphB) segregated as a single recessive trait for these mutants (Table 1). DNA was isolated from susceptible plants, and chromosome positions of the pbs mutations were established on the basis of linkage to CAPS (KONIECZNY and AUSUBEL 1993 ) and microsatellite (BELL and ECKER 1994 ) markers. Linkage data are shown in Table 3.

The pbs1-1 mutation did not map to a discrete location. We identified strong linkage to a region of ~40 cM on both chromosomes IV and V (Table 3). The pbs1-1 mutation was induced by fast neutrons, which are known to cause chromosome breaks. The lack of recombination seen on chromosomes IV and V could be explained by a translocation, accompanied by an inversion, between these chromosomes.

In contrast to pbs1-1, the pbs1-2 mutation, which was induced by EMS, mapped to a single region (Table 3). As shown in Figure 4, these data placed pbs1-2 on chromosome V between the markers nga249 and nga106. The genetic distances between markers was consistent with that derived from the Lister-Dean recombinant inbred lines, indicating no suppression of recombination in pbs1-2.

View larger version (16K): In this window In a new window Download PPT slide   Figure 4. Chromosome positions of the pbs1, pbs2, and pbs3 mutations. The map location of the pbs mutations relative to CAPS and microsatellite markers is shown. In parentheses is the position of each marker in centimorgans according to the Lister-Dean recombinant inbred map.

We determined that the pbs2 mutation was located on chromosome I. On the basis of recombination breakpoints, pbs2 was placed between the markers nga63 and NCC1 in a genetic interval of <0.4 cM (Table 3). RPS5, which confers resistance to DC3000(avrPphB), is also located near this region (Figure 4, SIMONICH and INNES 1995 ).

The pbs3 mutation exhibits partial dominance and maps to chromosome V:
When Col pbs3 plants were crossed to the Arabidopsis accessions Col-0, Col pbs1-1, Col pbs2, and Ler, all plants in the resulting F₁ generation appeared resistant to DC3000(avrPphB), indicating that pbs3 was recessive. However, segregation of the mutant trait in the F₂ generation of some of these crosses deviated significantly from expectations (Table 1). Susceptible plants were predominant in the F₂ generation resulting from the cross of Col pbs3 to Col pbs1-1. Assuming the pbs3 and pbs1-1 mutations are unlinked and recessive, the expected ratio of resistant to susceptible plants would be 9:7. We identified 13 resistant plants and 32 susceptible plants (² = 12.96), which is not statistically consistent with a 9:7 ratio. In this cross, the skewed segregation could result from a genetic interaction between the pbs1-1 and pbs3 alleles. Additionally, because pbs1 is linked to pbs3 (see below), the inversion/translocation that may be present in the pbs1-1 background could affect the segregation of pbs3 in this cross. Segregation of resistance in the backcross to Col-0 was consistent with a 3:1 ratio (69 resistant:34 susceptible; ² = 3.29), but the actual number of resistant to susceptible plants was closer to a 2:1 ratio. Segregation did not deviate significantly from 9:7 in the cross to Col pbs2, but only a few plants were assayed in the F₂ generation (Table 1). Taken together, we interpret these data to indicate that plants heterozygous for the pbs3 mutation may have slightly enhanced susceptibility that sometimes causes a susceptible phenotype, depending on genetic and/or environmental variables.

To map the pbs3 mutation, we used a cross to the Arabidopsis accession Ler. Because Ler lacks RPS5, we scored for the pbs3 mutant phenotype using DC3000 strains containing avrRpt2 or avrB rather than avrPphB. For DC3000(avrRpt2), segregation of resistant to susceptible plants was not consistent with a 3:1 ratio. Four hundred fifty plants were scored as resistant, and 198 plants were scored as susceptible (² = 10.67). Eighty-one DC3000(avrRpt2)-susceptible plants were tested initially for linkage, and the results suggested that pbs3 was located on chromosome V, near the marker nga249. Plants showing recombination near this region were retested for their response to DC3000(avrRpt2) in the F₃ generation. Fifteen plants (19%) segregated for resistance, indicating they were heterozygous for pbs3, and they were not included in the linkage data shown in Table 3.

An identical analysis was performed on DC3000 (avrB)-susceptible plants. Segregation was consistent with a single recessive gene (132 resistant:34 susceptible; ² = 1.80), but some susceptible plants were probably not identified because symptom development in response to DC3000(avrB) is weaker than that seen with DC3000(avrRpt2) (Figure 1). Because DC3000(avrRpt2) induces a stronger phenotype, it was used to infect recombinant plants in the F₃ generation. Of the 28 susceptible plants analyzed for linkage, 6 (21%) segregated for resistance and were not included in the data shown in Table 3.

After the elimination of plants that segregated for disease resistance in the F₃ generation, pbs3 was placed to a single genetic locus on chromosome V near the same region as pbs1-2 (Figure 4), between the markers nga249 and nga151. Recombination frequencies indicated pbs3 was ~0.6 cM from nga249 and 4.1 cM from nga151 (Table 3).

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have used a mutational approach to characterize molecular pathways leading to disease resistance in Arabidopsis. Three new genes were identified that exhibited susceptibility to several previously avirulent pathogens. The pbs1 mutation conferred full susceptibility to only one avirulent pathogen (Figure 1 and Figure 2, Table 2), indicating that this gene product may be critical to only one R gene-induced resistance pathway. The pbs2 and pbs3 mutant plants were susceptible to varying degrees against races of both prokaryotic and eukaryotic pathogens (Figure 1 and Figure 2, Table 2), suggesting that these two genes fulfill a function common to several R gene pathways.

PBS2 is genetically linked to RPS5, and PBS1 is linked to PBS3. The presence of functionally associated genes near the same genetic location has been observed before. For example, the R gene Pto is genetically linked to the Prf gene, which is required for Pto function (SALMERON et al. 1994 ). Similarly, two genes involved with self-incompatibility in Brassica are also tightly linked (BOYES and NASRALLAH 1993 ). For such examples, an argument can be made that the linked genes are both highly variable and specific to each other and, thus, must be inherited together to be functional. The PBS genes do not conform to this logic, however, as PBS2 and PBS3 mutations affect unlinked R genes.

PBS1 is closely associated with RPS5-mediated pathogen recognition:
Pathogen recognition mediated by R genes of the nucleotide-binding site (NBS)/leucine-rich repeat (LRR) class may require specific kinase partners (INNES 1995 ). To date, the involvement of both of these components has been shown only in the tomato for resistance mediated by the R gene Pto. Pto confers resistance against P. s. tomato carrying avrPto and encodes a functional serine/threonine kinase (RONALD et al. 1992 ; LOH and MARTIN 1995 ). Consistent with a postulated role as a receptor, Pto interacts with AvrPto in the yeast two-hybrid system (SCOFIELD et al. 1996 ; TANG et al. 1996 ).

In addition to Pto, recognition of AvrPto requires Prf, which encodes an NBS/LRR protein (SALMERON et al. 1996 ). Prf was originally identified in a screen for tomato mutants susceptible to P. s. tomato carrying avrPto (SALMERON et al. 1994 ). Mutations within Prf can fully suppress resistance mediated by Pto. An interaction between Pto and Prf has not been shown, but it has been suggested that the coupling of kinase and NBS/LRR components could confer specificity to a particular receptor complex (INNES 1995 ; SALMERON et al. 1996 ).

Similar to Prf and Pto, we have now identified two genes required for the recognition of avrPphB, RPS5, and PBS1. Since, like many R genes, RPS5 encodes an NBS/LRR protein (WARREN et al. 1998 ), a kinase is a candidate to be encoded by PBS1. Regardless of its structure, PBS1 is likely to be closely associated with the recognition of an avrPphB-derived elicitor because the pbs1 mutations fully suppressed disease resistance conferred by RPS5. Also, unlike most putative R gene signal transduction mutations that have been isolated, including pbs2 and pbs3, the pbs1 mutations did not greatly affect R genes other than RPS5 (Figure 1 and Figure 2, Table 2). A slight increase in sporulation was observed after inoculation with the P. parasitica isolates Emoy2 and Cand5 (Table 2), which may indicate that the PBS1 gene product can exhibit some specificity toward these RPP gene products.

PBS2 and NDR1 are involved in the same signal transduction pathways:
Like pbs2, the Arabidopsis mutation ndr1 affects R genes that specify resistance to avirulent P. s. tomato and P. parasitica. The pbs2 mutation appears to suppress the same set of R genes as ndr1. Both pbs2 and ndr1 exhibit increased disease symptoms in response to P. s. tomato carrying avrPphB, avrRpt2, and avrB (Figure 1; CENTURY et al. 1995 ). Additionally, neither mutation allows increased growth of P. s. tomato carrying avrRps4 (Figure 2; AARTS et al. 1998 ). Both pbs2 and ndr1 plants also show increased susceptibility to the same P. parasitica isolates (Table 2). For example, both mutants allow medium-to-high sporulation of Emoy2 and Cand5 on cotyledons and low sporulation of Hind4 and Wela3. The ndr1 mutation is not allelic to any of the pbs mutations, however, because they map to different chromosome locations (Table 3; CENTURY et al. 1995 ).

Given that the same R genes are affected by pbs2 and ndr1, these gene products may be closely associated with each other in the same signal transduction pathways. The precise role of NDR1 in pathogen resistance is currently unknown, as the NDR1 protein does not exhibit similarity to proteins of known function (CENTURY et al. 1997 ). However, NDR1 mRNA accumulation increases after infection by virulent or avirulent P. s. tomato strains and probably functions downstream of initial pathogen recognition.

Although the ndr1 and pbs2 mutations appear to affect the function of the same set of R genes, these gene products are not identical in their importance to all R gene signal transduction pathways. For example, NDR1 appears to be more critical than PBS2 for resistance specified by RPS2. Growth of P. s. tomato strain DC3000(avrRpt2) appears to be unrestricted in ndr1 leaves (CENTURY et al. 1995 ), but pbs2 plants were not fully susceptible to this pathogen (Figure 2). The pbs2 mutation also suppresses resistance against the P. parasitica isolate Wand1 much more strongly than does ndr1 (Table 2).

PBS3 acts to restrict growth of virulent and avirulent pathogens:
Col pbs3 mutant plants exhibited more severe disease symptoms in response to a virulent strain of P. s. tomato than did wild-type Col-0 plants (Figure 3). This enhanced disease susceptibility suggests that PBS3 is involved in controlling the growth of both virulent and avirulent pathogens.

Several Arabidopsis mutants have been isolated that show enhanced susceptibility to a virulent pathogen, and some of these mutations also affect resistance to avirulent pathogens (CAO et al. 1994 ; GLAZEBROOK and AUSUBEL 1994 ; DELANEY et al. 1995 ; GLAZEBROOK et al. 1996 , GLAZEBROOK et al. 1997 ). For example, eds1 mutants exhibit enhanced susceptibility to virulent P. parasitica and P. s. tomato (PARKER et al. 1996 ; AARTS et al. 1998 ). Additionally, eds1 mutations have been shown to disrupt resistance mediated by eight different RPP loci and a single bacterial resistance locus, RPS4. In contrast to eds1, the pbs3 mutation partially suppressed resistance conferred by four bacterial resistance genes, including RPS4, and allowed medium levels of sporulation on cotyledons by only one of the P. parasitica isolates that was tested and low sporulation by three other isolates (Figure 1 and Figure 2, Table 2). The Arabidopsis mutants pad1, pad2, pad4, and npr1 all show enhanced susceptibility to virulent P. s. maculicola (CAO et al. 1994 ; GLAZEBROOK and AUSUBEL 1994 ; GLAZEBROOK et al. 1996 , GLAZEBROOK et al. 1997 ). None of these mutants are altered in their response to avirulent bacteria, but their resistance is affected to different degrees against avirulent isolates of P. parasitica (DELANEY et al. 1995 ; GLAZEBROOK et al. 1997 ). These differences in phenotype suggest that PBS3 encodes a signaling component that is distinct from that encoded by EDS1, NPR1, or the PAD genes.

The phenotypes seen in Col pbs3 plants are reminiscent of plants with reduced levels of salicylic acid (SA). Transgenic Col-0 plants producing salicylate hydroxylase, which degrades SA, show enhanced susceptibility to virulent and avirulent P. s. tomato and P. parasitica pathogens (DELANEY et al. 1994 ). However, pbs3 does not seem to completely abolish SA-dependent defense responses because pbs3 plants are only partially suppressed in resistance to DC3000(avrRpt2) (Figure 2), while Col-0 plants expressing salicylate hydroxylase allow growth of DC3000(avrRpt2) equivalent to that seen in susceptible accessions of Arabidopsis (DELANEY et al. 1994 ). It is plausible that pbs3 plants have intermediate levels of SA.

PBS gene-dependent signal transduction pathways:
The identification of three new genes required by Arabidopsis to induce disease resistance allowed the dissection of signal transduction pathways that are activated by avirulent pathogens. A summary model based on assessment of R gene function disrupted by the pbs mutations and comparison to ndr1 and eds1 mutants is presented in Figure 5. In this model, the PBS1 gene product is closely associated with recognition of an avrPphB-derived elicitor, while the PBS2 and PBS3 gene products function downstream of pathogen recognition in multiple R gene defense pathways.

View larger version (21K): In this window In a new window Download PPT slide   Figure 5. Summary model of R gene disease-resistance pathways in Arabidopsis. The proposed placement of gene products in this pathway is based on phenotype analysis. These pathways are activated by an avr gene product derived from P. syringae. The recognition of this avr-based signal by RPS5 requires the PBS1 gene product. Similar proteins may be required by RPS2, RPM1, and RPS4, but they have not been identified. Potential roles of PBS2, PBS3, NDR1, and EDS1 in RPP gene-mediated resistance pathways are not shown.

The isolation of PBS1, which completely abolishes the function of a single R gene, RPS5, suggests that analogous genes could be identified by conducting mutant screens that assayed for loss of resistance conferred by other R genes. In Arabidopsis, however, genetic screens for loss of RPS2-, RPM1-, or RPP5-mediated disease resistance did not identify a mutation similar to pbs1 (KUNKEL et al. 1993 ; YU et al. 1993 ; BISGROVE et al. 1994 ; CENTURY et al. 1995 ; PARKER et al. 1996 ). This may suggest that PBS1 performs a function unique to the RPS5 disease-resistance pathway, or that redundant gene products perform its role in other R gene signal transduction pathways.

The simplest interpretation of the effects of the pbs2 and pbs3 mutation suggests that PBS3 operates downstream of PBS2 (Figure 5). It is also possible, however, that PBS3 is involved in an independent pathway that contributes to resistance. The enhanced susceptibility to virulent pathogens and the failure of the pbs3 mutation to completely abolish disease resistance supports this idea. Continued analysis of the pbs mutants and characterization of the corresponding gene products should further our understanding of the processes used by plants to limit pathogen growth.

	ACKNOWLEDGMENTS

We thank Dr. Mark Estelle for providing mutagenized Arabidopsis seed and John Danzer, Sandra Szerszen, Patricia Mowery, and Anna Bocian for technical assistance. Research in Indiana was supported by grant R01 GM46451 from the Institute of General Medical Sciences of the National Institutes of Health to R.W.I. Research at Horticulture Research International was supported by core funding to E.H. from the Biotechnology and Biological Systems Research Council.

Manuscript received October 27, 1998; Accepted for publication January 20, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AARTS, N., M. METZ, E. HOLUB, B. J. STASKAWICZ, and M. J. DANIELS et al., 1998  Different requirements for EDS1 and NDR1 by disease resistance genes define at least two R gene-mediated signalling pathways in Arabidopsis. Proc. Natl. Acad. Sci. USA 95:10306-10311[Abstract/Free Full Text].

BAKER, B., P. ZAMBRYSKI, B. STASKAWICZ, and S. P. DINESH-KUMAR, 1997  Signaling in plant-microbe interactions. Science 276:726-733[Abstract/Free Full Text].

BELL, C. J. and J. R. ECKER, 1994  Assignment of 30 microsatellite loci to the linkage map of Arabidopsis. Genomics 19:137-144[Medline].

BENT, A. F., 1996  Plant disease resistance genes: function meets structure. Plant Cell 8:1757-1771[Medline].

BISGROVE, S. R., M. T. SIMONICH, N. M. SMITH, A. SATTLER, and R. W. INNES, 1994  A disease resistance gene in Arabidopsis with specificity for two different pathogen avirulence genes. Plant Cell 6:927-933[Abstract].

BOYES, D. C. and J. B. NASRALLAH, 1993  Physical linkage of the SLG and SRK genes at the self-incompatibility locus of Brassica oleracea.. Mol. Gen. Genet. 236:369-373[Medline].

CAO, H., S. A. BOWLING, S. GORDON, and X. DONG, 1994  Characterization of an Arabidopsis mutant that is nonresponsive to inducers of systemic acquired resistance. Plant Cell 6:1583-1592[Abstract].

CENTURY, K. S., E. B. HOLUB, and B. J. STASKAWICZ, 1995  NDR1, a locus of Arabidopsis thaliana that is required for disease resistance to both a bacterial and a fungal pathogen. Proc. Natl. Acad. Sci. USA 92:6597-6601[Abstract/Free Full Text].

CENTURY, K. S., A. D. SHAPIRO, P. P. REPETTI, D. DAHLBECK, and E. HOLUB et al., 1997  NDR1, a pathogen-induced component required for Arabidopsis disease resistance. Science 278:1963-1965[Abstract/Free Full Text].

DANGL, J. L., E. B. HOLUB, T. DEBENER, H. LEHNACKERS, C. RITTER et al., 1992 Genetic definition of loci involved in Arabidopsis-pathogen interactions, pp. 393–418 in Methods in Arabidopsis Research, edited by C. KONCZ, N.-H. CHUA and J. SCHELL. World Scientific Publishing Co., London.

DELANEY, T. P., S. J. UKNES, B. VERNOOIJ, L. FRIEDRICH, and K. WEYMANN et al., 1994  A central role of salicylic acid in plant disease resistance. Science 266:1247-1249[Abstract/Free Full Text].

DELANEY, T. P., L. FRIEDRICH, and J. A. RYALS, 1995  Arabidopsis signal transduction mutant defective in chemically and biologically induced disease resistance. Proc. Natl. Acad. Sci. USA 92:6602-6606[Abstract/Free Full Text].

ELLIS, J. and D. JONES, 1998  Structure and function of proteins controlling strain-specific pathogen resistance in plants. Curr. Opin. Plant Biol. 1:288-293[Medline].

FLOR, H., 1971  Current status of the gene-for-gene concept. Annu. Rev. Phytopathol. 9:275-296.

FREIALDENHOVEN, A., B. SCHERAG, K. HOLLRICHER, D. B. COLLINGE, and H. THORDAL-CHRISTENSEN et al., 1994  Nar-1 and Nar-2, two loci required for Mla₁₂-specified race-specific resistance to powdery mildew in barley. Plant Cell 6:983-994[Abstract].

FRYE, C. A. and R. W. INNES, 1998  An arabidopsis mutant with enhanced resistance to powdery mildew. Plant Cell 10:947-956[Abstract/Free Full Text].

GABRIEL, D. W. and B. G. ROLFE, 1990  Working models of specific recognition in plant-microbe interactions. Annu. Rev. Phytopathol. 28:365-391.

GLAZEBROOK, J. L. and F. M. AUSUBEL, 1994  Isolation of phytoalexin-deficient mutants of Arabidopsis thaliana and characterization of their interactions with bacterial pathogens. Proc. Natl. Acad. Sci. USA 91:8955-8959[Abstract/Free Full Text].

GLAZEBROOK, J., E. E. ROGERS, and F. M. AUSUBEL, 1996  Isolation of Arabidopsis mutants with enhanced disease susceptibility by direct screening. Genetics 143:973-982[Abstract].

GLAZEBROOK, J., M. ZOOK, F. MERT, I. KAGAN, and E. E. ROGERS et al., 1997  Phytoalexin-deficient mutants of Arabidopsis reveal that PAD4 encodes a regulatory factor and that four PAD genes contribute to downy mildew resistance. Genetics 146:381-392[Abstract].

HAMMOND-KOSACK, K. E. and J. D. G. JONES, 1996  Resistance gene-dependent plant defense responses. Plant Cell 8:1773-1791[Medline].

HAMMOND-KOSACK, K. E., D. A. JONES, and J. D. G. JONES, 1994  Identification of two genes required in tomato for full Cf-9-dependent resistance to Cladosporium fulvum.. Plant Cell 6:361-374[Abstract].

HINSCH, M. and B. J. STASKAWICZ, 1996  Identification of a new Arabidopsis disease resistance locus, RPS4, and cloning of the corresponding avirulence gene, avrRps4, from Pseudomonas syringae pv. pisi. Mol. Plant-Microbe Interact. 9:55-61[Medline].

HOLUB, E. B., 1997 Organization of resistance genes in Arabidopsis, pp. 5–26 in The Gene-for-Gene Relationship in Plant Parasite Interactions, edited by I. R. CRUTE, E. B. HOLUB, and J. J. BURDON. CAB International, Wallingford, UK.

HOLUB, E. B., J. L. BEYNON, and I. R. CRUTE, 1994  Phenotypic and genotypic characterization of interactions between isolates of Peronospora parasitica and accessions of Arabidopsis thaliana.. Mol. Plant-Microbe Interact. 7:223-239.

INNES, R., 1995  Plant-parasite interactions: has the gene-for-gene model become outdated? Trends. Microbiol. 3:483-485[Medline].

INNES, R. W., 1998  Genetic dissection of R gene signal transduction pathways. Curr. Opin. Plant Biol. 1:299-304[Medline].

INNES, R. W., S. R. BISGROVE, N. M. SMITH, A. F. BENT, and B. J. STASKAWICZ et al., 1993  Identification of a disease resistance locus in Arabidopsis that is functionally homologous to the Rpg1 locus of soybean. Plant J. 4:813-820[Medline].

JONES, D. A. and J. D. G. JONES, 1997  The role of leucine-rich repeat proteins in plant defences. Adv. Bot. Res. 24:89-167.

JORGENSEN, J. H., 1988  Genetic analysis of barley mutants with modifications of powdery mildew resistance gene Mla-12.. Genome 30:129-132.

JORGENSEN, J. H., 1996  Effect of three suppressors on the expression of powdery mildew resistance genes in barley. Genome 39:492-498.

KONIECZNY, A. and F. M. AUSUBEL, 1993  A procedure for mapping Arabidopsis mutations using co-dominant ecotype-specific PCR-based markers. Plant J. 4:403-410[Medline].

KOSAMBI, D. D., 1944  The estimation of map distances from recombination values. Ann. Eugen. 12:172-175.

KUNKEL, B. N., 1996  A useful weed put to work: genetic analysis of disease resistance in Arabidopsis thaliana.. Trends Genet. 12:63-69[Medline].

KUNKEL, B. N., A. F. BENT, D. DAHLBECK, R. W. INNES, and B. J. STASKAWICZ, 1993  RPS2, an Arabidopsis disease resistance locus specifying recognition of Pseudomonas syringae strains expressing the avirulence gene avrRpt2.. Plant Cell 5:865-875[Abstract/Free Full Text].

LOH, Y. T. and G. B. MARTIN, 1995  The Pto bacterial resistance gene and the Fen insecticide sensitivity gene encode functional protein kinases with serine/threonine specificity. Plant Physiol. 108:1735-1739[Abstract].

PARKER, J. E., E. B. HOLUB, L. N. FROST, A. FALK, and N. D. GUNN et al., 1996  Characterization of eds1, a mutation in Arabidopsis suppressing resistance to Peronospora parasitica specified by several different RPP genes. Plant Cell 8:2033-2046[Abstract].

RONALD, P. C., J. M. SALMERON, F. M. CARLAND, and B. J. STASKAWICZ, 1992  The cloned avirulence gene avrPto induces disease resistance in tomato cultivars containing the Pto resistance gene. J. Bacteriol. 174:1604-1611[Abstract/Free Full Text].

SALMERON, J. M., S. J. BARKER, F. M. CARLAND, A. Y. MEHTA, and B. J. STASKAWICZ, 1994  Tomato mutants altered in bacterial disease resistance provide evidence for a new locus controlling pathogen recognition. Plant Cell 6:511-520[Abstract].

SALMERON, J. M., G. E. D. OLDROYD, C. M. T. ROMMENS, S. R. SCOFIELD, and H.-S. KIM et al., 1996  Tomato Prf is a member of the leucine-rich repeat class of plant disease resistance genes and lies embedded within the Pto kinase gene cluster. Cell 86:123-134[Medline].

SCOFIELD, S. R., C. M. TOBIAS, J. P. RATHJEN, J. H. CHANG, and D. T. LAVELLE et al., 1996  Molecular basis of gene-for-gene specificity in bacterial speck disease of tomato. Science 274:2063-2065[Abstract/Free Full Text].

SIMONICH, M. T. and R. W. INNES, 1995  A disease resistance gene in Arabidopsis with specificity for the avrPph3 gene of Pseudomonas syringae pv. phaseolicola. Mol. Plant-Microbe Interact. 8:637-640[Medline].

TANG, X., R. D. FREDERICK, J. ZHOU, D. A. HALTERMAN, and Y. JIA et al., 1996  Initiation of plant disease resistance by physical interaction of AvrPto and Pto kinase. Science 274:2060-2063[Abstract/Free Full Text].

TORP, J. and J. JORGENSEN, 1986  Modification of barley powdery mildew resistance gene Mla-12 by induced mutation. Can. J. Genet. Cytol. 28:725-731.

WARREN, R. F., A. HENK, P. MOWERY, E. HOLUB, and R. W. INNES, 1998  A mutation within the leucine-rich repeat domain of the Arabidopsis disease resistance gene RPS5 partially suppresses multiple bacterial and downy mildew resistance genes. Plant Cell 10:1439-1452[Abstract/Free Full Text].

WHALEN, M. C., R. W. INNES, A. F. BENT, and B. J. STASKAWICZ, 1991  Identification of Pseudomonas syringae pathogens of Arabidopsis and a bacterial locus determining avirulence on both Arabidopsis and soybean. Plant Cell 3:49-59[Abstract/Free Full Text].

YU, G.-L., F. KATAGIRI, and F. M. AUSUBEL, 1993  Arabidopsis mutations at the RPS2 locus result in loss of resistance to Pseudomonas syringae strains expressing the avirulence gene avrRpt2.. Molec. Plant-Microbe Interact. 6:434-443.