Isolation of New Arabidopsis Mutants With Enhanced Disease Susceptibility to Pseudomonas syringae by Direct Screening

Corresponding author: Frederick M. Ausubel, Department of Molecular Biology, Massachusetts General Hospital, Boston, MA 02114, ausubel{at}frodo.mgh.harvard.edu (E-mail).

Communicating editor: V. SUNDARESAN

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

To identify plant defense components that are important in restricting the growth of virulent pathogens, we screened for Arabidopsis mutants in the accession Columbia (carrying the transgene BGL2-GUS) that display enhanced disease susceptibility to the virulent bacterial pathogen Pseudomonas syringae pv. maculicola (Psm) ES4326. Among six (out of a total of 11 isolated) enhanced disease susceptibility (eds) mutants that were studied in detail, we identified one allele of the previously described npr1/nim1/sai1 mutation, which is affected in mounting a systemic acquired resistance response, one allele of the previously identified EDS5 gene, and four EDS genes that have not been previously described. The six eds mutants studied in detail (npr1-4, eds5-2, eds10-1, eds11-1, eds12-1, and eds13-1) displayed different patterns of enhanced susceptibility to a variety of phytopathogenic bacteria and to the obligate biotrophic fungal pathogen Erysiphe orontii, suggesting that particular EDS genes have pathogen-specific roles in conferring resistance. All six eds mutants retained the ability to mount a hypersensitive response and to restrict the growth of the avirulent strain Psm ES4326/avrRpt2. With the exception of npr1-4, the mutants were able to initiate a systemic acquired resistance (SAR) response, although enhanced growth of Psm ES4326 was still detectable in leaves of SAR-induced plants. The data presented here indicate that eds genes define a variety of components involved in limiting pathogen growth, that many additional EDS genes remain to be discovered, and that direct screens for mutants with altered susceptibility to pathogens are helpful in the dissection of complex pathogen response pathways in plants.

PLANTS are constantly challenged by a variety of potentially pathogenic microorganisms and have developed an array of constitutive and inducible defense strategies designed to prevent and limit infections (LAMB et al. 1989 ; LAMB 1994 ). A major focus of biochemical and molecular research in plant pathology has been the elucidation of the molecular mechanisms underlying pathogen recognition and the signaling cascades leading to the activation of defense responses. Resistance to so-called avirulent pathogens is often triggered by the specific recognition of a pathogen-derived signal [generated by an avirulence (avr) gene] which is thought to be recognized by a corresponding host receptor [encoded by a resistance (R) gene] (FLOR 1971 ). In many cases, avr-R-mediated resistance is accompanied by rapid programmed host cell death at the site of infection (called the hypersensitive response or HR) and other defense-related responses, such as cell wall reinforcement, production of reactive oxygen species and antimicrobial compounds (phytoalexins), and accumulation of pathogenesis-related (PR) proteins (KAUFFMANN et al. 1987 ; LEGRAND et al. 1987 ; LAMB et al. 1989 ; DIXON and LAMB 1990 ; PONSTEIN et al. 1994 ; JACH et al. 1995 ; HAMMOND-KOSACK and JONES 1996 ).

Many of the same host defense responses that are involved in avr-R gene mediated resistance also play a major role in defending the host against so-called virulent pathogens that elicit prominent disease symptoms (DIXON and LAMB 1990 ; DIXON et al. 1994 ). In general, however, infection with an avirulent pathogen (that leads to a so-called incompatible interaction) results in more rapid or extensive induction of host responses than infection with a virulent pathogen (that leads to a so-called compatible interaction). In either case, host defense responses restrict the growth and spread of the pathogen; but relatively little is known about the actual contribution of particular responses to limit pathogen growth or the signal transduction pathways that mediate their expression.

Some host defense-related genes encode proteins with enzymatic properties that suggest they play a direct role in conferring resistance to pathogens. The PR proteins ß-1,3-glucanase and chitinase, for example, are lytic enzymes that are potentially capable of degrading polysaccharides found in the cell wall of many fungi, and both have been shown to exhibit anti-fungal properties when assayed in vitro (SCHLUMBAUM et al. 1986 ; MAUCH et al. 1988 ; SELA-BUURLAGE et al. 1993 ; MELCHERS et al. 1994 ; PONSTEIN et al. 1994 ; NIDERMAN et al. 1995 ). Other studies have revealed that constitutive expression of defense-related genes or certain phytoalexin biosynthetic genes in transgenic plants can result in enhanced pathogen resistance (BROGLIE et al. 1991 ; ALEXANDER et al. 1993 ; HAIN et al. 1993 ; LIU et al. 1994 ; ZHU et al. 1994 ; TERRAS et al. 1995 ).

A large number of defense responses have been identified on the basis of observations that they are induced in response to pathogen attack. In the past, however, most host-pathogen systems studied by plant pathologists were not amenable to high throughput genetic analysis, and it was therefore difficult to assess the significance of specific defense responses. To overcome this problem, many recent host-pathogen studies have utilized Arabidopsis thaliana as a model host plant, and comprehensive genetic analysis of host pathogen interactions is now feasible (DANGL 1993 ; CRUTE et al. 1994 ; CRUTE and PINK 1996 ; KUNKEL 1996 ). In brief, the isolation and characterization of a variety of Arabidopsis defense response mutants has helped to dissect the processes involved in combating specific pathogens and has uncovered new defense mechanisms and signal transduction pathways not previously correlated with a known biochemical response (for reviews see GLAZEBROOK et al. 1997A ; YANG et al. 1997 ).

Two examples of the plant defense response that have been subjected to genetic analysis in Arabidopsis are phytoalexin synthesis and systemic acquired resistance (SAR). A series of phytoalexin deficient (pad) mutants that synthesize reduced levels of camalexin in response to the bacterial pathogen Pseudomonas syringae pv. maculicola (Psm) ES4326 have been isolated (GLAZEBROOK and AUSUBEL 1994 ; GLAZEBROOK et al. 1997B ). When infected with a low dose of Psm ES4326 or the fungal pathogen Peronospora parasitica, certain pad mutants exhibit an enhanced disease susceptibility (eds) phenotype (GLAZEBROOK and AUSUBEL 1994 ; GLAZEBROOK et al. 1997B ).

Three allelic Arabidopsis mutants that affect SAR are npr1 (CAO et al. 1994 ; CAO et al. 1997 ), nim1 (DELANEY et al. 1995 ; RYALS et al. 1997 ) and sai1 (SHAH et al. 1997 ). SAR is initiated upon infection of a plant with a necrotizing pathogen and leads to the development of systemic resistance to various pathogens throughout the plant (ENYEDI et al. 1992A ; MALAMY and KLESSIG 1992 ; RYALS et al. 1996 ). SAR is correlated with the accumulation of PR proteins in uninfected tissue and an SAR-like response can be elicited by exogenous treatment of plants with salicylic acid (SA) (GAFFNEY et al. 1993 ). The npr1/sai1/nim1 mutants do not activate PR gene expression after treatment with SA, fail to become systemically resistant in response to infection by a necrotizing pathogen, and display increased susceptibility to P. syringae and P. parasitica. Like some of the pad mutants, npr1/sai1/nim1 mutants exhibit an enhanced disease susceptibility phenotype (CAO et al. 1994 ; DELANEY et al. 1995 ; SHAH et al. 1997 ).

Based on the eds phenotypes of pad and npr1 mutants, a direct screen for Arabidopsis eds mutants was carried out previously in our laboratory with the goal of identifying genes other than NPR1 and PAD genes that are involved in restricting the growth of the virulent pathogen Psm ES4326. This screen led to the identification of 10 eds mutants that are neither pad nor npr1 mutants that define at least eight new complementation groups (GLAZEBROOK et al. 1996 ; ROGERS and AUSUBEL 1997 ). Detailed characterization of four of these eds mutants (eds5-1, eds6-1, eds7-1, and eds9-1) indicated that these EDS genes define a diverse set of previously unknown defense-related functions limiting the severity of virulent bacterial infections (ROGERS and AUSUBEL 1997 ). Because of the low frequency of allelic pairs of eds mutants recovered in the screen, we hypothesized that the screen was not saturating and that many more EDS genes remain to be discovered.

In this article, we report the results of a new screen for eds mutants to low dose infections of Psm ES4326. Among 11 putative eds mutants isolated, we identified one npr1 mutant (CAO et al. 1997 ) , one mutant that is allelic to the previously isolated mutant eds5-1 (GLAZEBROOK et al. 1996 ; ROGERS and AUSUBEL 1997 ), and mutants with defects in at least four other EDS genes not previously described.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Bacteria, plants, growth conditions:
The bacterial strains Pseudomonas syringae pv. maculicola ES4326 (DONG et al. 1991 ), P. s. pv. tomato DC3000 (CUPPELS 1986 ), P. s. pv. phaseolicola 3121 (RAHME et al. 1992 ), Pseudomonas aeruginosa UCBPP-PA14 (RAHME et al. 1995 ) and Xanthomonas campestris pv. raphani 1946 (PARKER et al. 1993 ) have been described previously. The avirulent strains carry avrRpt2 on the plasmid pLH12 (WHALEN et al. 1991 ). P. syringae strains were grown at 28° in King's B medium (10 mg/ml protease peptone, 2 mg/ml K₂HPO₄, 10 mg/ml glycerol, 6 mM MgSO₄, pH 7.0) supplemented with 100 µg/ml streptomycin for P. s. maculicola ES4326, 50 µg/ml rifampicin for P. s. tomato DC3000 and P. s. phaseolicola 3121, or 10 µg/ml tetracycline for strains carrying pLH12. P. aeruginosa UCBPP-PA14 was grown in Luria-Bertani medium (Difco, Detroit) supplemented with 50 µg/ml rifampicin at 37°. X. campestris pv. raphani 1946 was cultured at 28° in NYG medium containing 50 µg/ml rifampicin.

Transgenic Arabidopsis thaliana ecotype Columbia containing a BGL2-GUS construct (DONG et al. 1991 ) was grown in Metromix 2000 soil (Scott, Marysville, OH), either in a climate-controlled greenhouse (20 ± 2°, relative humidity 70 ± 5%) on a 12-hr-light/12-hr-dark cycle or in a Conviron growth chamber (Controlled Environments Ltd., Winnipeg, Manitoba, Canada) (20 ± 2°, 90% relative humidity, 150 µE m^-2 sec^-1 fluorescent illumination) on a 12-hr-light/12-hr-dark cycle. The initial screen for mutants exhibiting enhanced susceptibility to Psm ES4326 was performed in a greenhouse; all other experiments were conducted in the Conviron growth chamber. Plants infected with P. aeruginosa UCBPP-PA14 were transferred into a Percival growth chamber (Boone, IA) at 30° and 100% humidity (RAHME et al. 1995 ). Plants were infected with suspensions of bacterial cells in 10 mM MgSO₄ by means of a 1-ml syringe (without a needle) forcing the suspension through the stomata by gently pressing the syringe against the abaxial side of the leaves.

Mutant screen and genetic crosses:
The preparation of EMS mutagenized seeds of the Arabidopsis BGL2-GUS transgenic line has previously been described (CAO et al. 1994 ). Plants grown in the greenhouse were infected with Psm ES4326 at a dose of 10⁴ cfu/cm² (which is equivalent to OD₆₀₀ of 0.002), and disease symptoms were examined visually three days after infection. Genetic crosses were performed by dissecting immature flowers of the pollen recipient prior to anther dehiscence and applying pollen from the donor to the recipient pistil. Segregation in the F₂ generation was analyzed with a ² test for goodness of fit.

Bacterial growth assays:
Plants grown in a growth chamber were infected with the relevant bacterium at the dose indicated. Three days later, bacterial growth was assayed by excising a leaf sample consisting of two 0.13-cm² disks from each infected leaf using a cork borer, grinding the sample in 10 mM MgSO₄ using a plastic pestle, and plating appropriate dilutions on the relevant bacterial growth medium containing the appropriate antibiotic(s). Four to eight leaves were excised per treatment per genotype. Growth data are reported as means and standard deviations of the log of the number of colony-forming bacterial units per square centimeter of leaf area (log cfu/cm²).

Assays for BGL2-GUS reporter gene activity after SA treatment:
Seeds were germinated and grown for 10 days on Murashige and Skoog (MS) medium (GIBCO BRL, Gaithersburg, MD) supplemented with 0.6% agar and 2% sucrose under 50 µE M^-2 sec^-1 fluorescent illumination at 20°. Thereafter, the seedlings were transferred either to fresh MS medium or to MS medium containing 0.5 mM sodium salicylate. After another 7 days of growth, a single leaf per seedling was removed and the assay for ß-glucuronidase (GUS) activity was performed as described previously (CAO et al. 1994 ).

Camalexin assay:
Four-week-old plants were inoculated with Psm ES4326 at a dose of 10⁶ cfu/cm² and 36 hr later camalexin was determined as described previously (GLAZEBROOK et al. 1996 ).

Powdery mildew infections and infection rating (IR) scores:
Erysiphe orontii isolate MGH was grown and maintained on the Arabidopsis mutant pad4 (GLAZEBROOK and AUSUBEL 1994 ; GLAZEBROOK et al. 1997A ) which supports more growth of this fungus (T. L. REUBER and F. M. AUSUBEL, unpublished data). Dry conidia derived from 2- to 3-wk-old infections of A. thaliana plants were used as inoculum. Four- to five-week-old plants were infected using modified settling towers (ADAM and SOMERVILLE 1996 ) consisting of a metal tower (14 cm x 14 cm x 71 cm) covered with nylon mesh (95-µm openings) (Small Parts, Inc., Miami Lakes, FL). Conidia from one heavily infected leaf were brushed onto the nylon mesh and forced through to break up the conidial chains. Infected plants were examined visually for the appearance of white mats of mycelium and conidiophores at 7 days postinoculation (dpi) and at 16 dpi. The IR scores were evaluated by comparison with the level of infection observed on Col-0 wild-type plants. The level of fungal infection was ranked on a scale from 0 to "++++", with a score of 0 indicating that no visible symptoms of infection were apparent on the leaves. A score of "+" indicated that only singular spots of powdery mildew were apparent by eye. An intermediate density of fungal growth with approximately 20% of the leaf blade covered by a white mat of mycelium and conidiophores was scored as "++", and a score of "+++" indicated that the mycelial mat covered about 50% of the leaf (or bolt) surface. An IR score of "++++" was assigned to plants supporting extremely dense fungal growth (the mycelium covering the whole surface of the leaf blade apparent as a dense white mat).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Isolation of Arabidopsis mutants with enhanced disease susceptibility in response to Psm ES4326 infection:
Genetic analysis of 12 Arabidopsis eds mutants isolated previously in our laboratory showed that they defined eight complementation groups, indicating that the screen for eds mutants had not been saturated. To obtain additional Arabidopsis mutants that are altered in their response to the virulent bacterial phytopathogen P. syringae pv. maculicola (Psm) ES4326, we performed a new genetic screen using plants derived from EMS-mutagenesis of BGL2-GUS transgenic seeds (Columbia ecotype) (DONG et al. 1991 ). BGL2 (also known as PR2) encodes a ß-1,3-glucanase and the chimeric BGL2-GUS gene serves as a reporter of the plant defense response to pathogen attack. Two leaves of each of 3,460 M₂ plants were inoculated with Psm ES4326 at a dose of 10⁴ cfu/cm² leaf area, which is not sufficient to elicit symptoms in wild-type plants, and the development of chlorotic spots on infected leaves was scored by visual examination. A total of 145 candidate eds mutants that displayed more extensive chlorosis in comparison to wild-type plants were identified and were retested in the M₃ and M₄ generations using the quantitative Psm ES4326 growth assay described in MATERIALS AND METHODS. As illustrated in Figure 1 and summarized in Table 1 of the 145 candidate eds mutants reproducibly displayed enhanced chlorosis and allowed at least 10-fold more growth of Psm ES4326 in comparison to wild-type plants. These 11 putative eds mutants were saved and were subjected to additional analysis described below.

View larger version (94K): In this window In a new window Download PPT slide   Figure 1. —Psm ES4326 elicits more severe symptoms in eds mutants than it does in wild-type plants. Three days after infection with Psm ES4326 at a dose of 103 cfu/cm2, leaves were excised and photographed. The left panel shows a wild-type leaf, the right panels show one leaf each of mutants eds-15, eds-16, and eds-17.

One of the eds mutants is an allele of npr1:
Previously identified categories of Arabidopsis mutants that allow more growth of P. syringae include npr1 (CAO et al. 1994 ) which is also called nim1 (DELANEY et al. 1995 ) or sai1 (SHAH et al. 1997 ), certain pad mutants (GLAZEBROOK and AUSUBEL 1994 ; GLAZEBROOK et al. 1997B ), and previously isolated eds mutants (GLAZEBROOK et al. 1996 ; ROGERS and AUSUBEL 1997 ). The defining features of npr1 mutants, in addition to their enhanced susceptibility phenotype, is the failure to activate expression of the defense-related genes BGL2, PR1, and PR5 following treatment with SA and the inability to mount an SAR response after infection with a necrotizing pathogen. Because the eds mutants isolated in this screen carry the SA-responsive promoter region of the BGL2 gene fused to the GUS coding region (DONG et al. 1991 ), we compared the induction of the BGL2-GUS reporter gene in wild-type and eds mutant seedlings grown in the presence or absence of SA as described in MATERIALS AND METHODS. As summarized in Table 1, wild-type plants and all of the eds mutants exhibited approximately the same level of BGL2-GUS activity in the presence of SA except eds-17 and eds-215 which had no detectable GUS activity and approximately 25% of wild-type GUS activity, respectively. These data indicated that eds-17 and eds-215 might contain mutant npr1 alleles. In genetic complementation analysis when eds-17 and eds-215 were crossed to the well-characterized npr1 mutant npr1-3, eds-17 failed to complement the Psm ES4326 growth and enhanced disease symptoms phenotypes of npr1-3 whereas eds-215 fully complemented these phenotypes.

NPR1 has recently been cloned using a map-based approach and was found to encode a novel protein containing ankyrin repeats (CAO et al. 1997 ). Previously reported sequence analysis revealed that eds-17 carries a single base-pair change in the NPR1 gene disrupting the splice site of the third intron junction (CAO et al. 1997 ). As a result of these studies and because the eds phenotype of eds-17 segregates 1:3 in the F₂ generation when crossed to wild type (see below), the mutation in eds-17 was renamed npr1-4.

Arabidopsis phytoalexin deficient (pad) mutants are defective in the accumulation of camalexin and some pad mutants allow more growth of Psm ES4326 (GLAZEBROOK and AUSUBEL 1994 ; GLAZEBROOK et al. 1997B ). We therefore assayed the accumulation of camalexin in the eds mutants after infection with Psm ES4326 by visual examination of camalexin fluorescence in a biochemical assay (see MATERIALS AND METHODS). As summarized in Table 1, none of the eleven eds mutants tested had significantly different camalexin levels than wild-type plants.

Genetic analyses of eds mutants:
We chose npr1-4, eds-15, eds-16, eds-19, eds-111, and eds-215 for additional genetic analysis because they reproducibly displayed the most severe eds phenotypes of the eleven mutants isolated in this screen. These six eds mutants were first subjected to complementation analysis (essentially as described by GLAZEBROOK et al. 1996 ) by crossing to each other and to wild-type Columbia plants and in each resulting F₁ generation, three to eight plants were assayed for the development of enhanced chlorosis and growth of Psm ES4326. Complementation was defined as the absence of an eds phenotype in F₁ plants in response to Psm ES4326 infection. As shown in Table 2, all six of these eds mutations were recessive and complemented each other, in support of the conclusion reached above that eds-215 is not an npr1 mutant.

Previous genetic studies in our laboratory have identified eight EDS loci, designated EDS2, EDS3, EDS4, EDS5, EDS6, EDS7, EDS8, and EDS9 (GLAZEBROOK et al. 1996 ; ROGERS and AUSUBEL 1997 ). To test whether the five eds mutants isolated in this study (excluding npr1-4) define new complementation groups, we extended the complementation tests including the previously isolated eds mutants. The five eds mutants (eds-15, eds-16, eds-19, eds-111, and eds-215) were crossed to alleles of the previously isolated eds mutants (eds2-1, eds3-1, eds4-1, eds5-1, eds6-1, eds7-1, eds8-1, and eds9-1) in pairwise combination. Three days after inoculation with Psm ES4326, symptom severity and Psm ES4326 growth were examined in the resulting F₁ generation plants, and in wild-type and mutant control plants. Also included in this complementation analysis was the mutant eds1-2, that carries a mutation in a recessive gene required for race-specific resistance to several Peronospora parasitica isolates (PARKER et al. 1996 ). Under the conditions assayed in our laboratory, eds1-2 allows fivefold more growth of Psm ES4326 compared to wild type (data not shown). Table 3 shows that among the eds mutants tested only eds-19 and eds5-1 failed to complement each other. The results of the Psm ES4326 growth assay for six F₁ cross progeny of eds5-1 x eds-19 as well as for wild-type and mutant control plants are shown in Table 4. We confirmed that the F₁ plants derived from this cross were true cross-progeny because they did not display the long trichome phenotype of eds-19 (unlinked recessive mutation) and they did not display the fluorescence phenotype of fah1-2 homozygotes (genetic background of eds5-1). In the F₂ generation, 73 out of 73 progeny plants derived from a single F₁ parent that were tested displayed an enhanced disease susceptibility phenotype upon challenge with a low dose of Psm ES4326 and 19 out of 73 (approximately one quarter of the F₂ plants tested) displayed long trichomes, a phenotype derived from the pollen donor eds-19.

Based on the results of this genetic analysis, the names of the five eds mutants isolated in this study were changed as follows: eds-15 to eds10-1, eds-16 to eds11-1, eds-19 to eds5-2, eds-111 to eds12-1, and eds-215 to eds13-1. In summary, this screen led to the identification of at least four new eds loci involved in restricting the growth of Psm ES4326, plus two alleles of two previously known genes that affect disease susceptibility, eds5-2 and npr1-4.

The segregation patterns of the eds phenotypes reveal mutations in single nuclear genes:
In the crosses between Col-0 parent and eds mutant plants (Table 2), F₁ plants displayed symptoms and bacterial titers similar to wild type in response to infection with Psm ES4326. This result indicates that all six eds mutations are recessive. Selfing of the F₁ generation plants produced F₂ progeny that were analyzed for the segregation of the eds phenotype in response to Psm ES4326 infection at a low dose (10³ cfu/cm²). F₂ plants displaying mild disease symptoms resembling those of Col-0 plants were scored as Eds⁺, and F₂ plants displaying severe symptoms resembling those of eds parent plants were scored as Eds^-. The results shown in Table 5 are consistent with the hypothesis that the eds phenotypes of npr1-4, eds5-2, eds10-1, eds11-1, eds12-1, and eds13-1 segregate in a 3:1 ratio and therefore result from recessive mutations in single nuclear genes. Interestingly, all eds mutants characterized to date seem to be recessive alleles of single nuclear genes. The phenotypic characterization described below was performed with plants that had been taken through at least one parental backcross.

Response to the bacterial avirulence gene avrRpt2 is unaffected in the eds mutants:
We have shown that the six eds mutants (npr1-4, eds5-2, eds10-1, eds11-1, eds12-1, and eds13-1) allow at least 10-fold more growth of the virulent pathogen Psm ES4326 as compared to wild-type plants (Figure 2 and Table 1). It was of interest to determine whether the mutants are also compromised in the response to avirulent bacteria, such as the isogenic avirulent strain Psm ES4326 carrying the avirulence gene avrRpt2. P. syringae strains carrying avrRpt2 elicit a rapid gene-for-gene resistance response in Col-0 plants, the so-called hypersensitive response (HR), mediated by the resistance gene RPS2. As a consequence, bacteria carrying avrRpt2 reach a lower density in infected leaves than isogenic strains lacking avrRpt2. Leaves of Col-0 wild-type and eds mutant plants were inoculated with the virulent strain Psm ES4326 or with the isogenic avirulent strain Psm ES4326/avrRpt2 at a dose of 10³ cfu/cm². Figure 2 shows that wild-type plants and the six eds mutants npr1-4, eds5-2, eds10-1, eds11-1, eds12-1, and eds13-1 were not significantly altered compared to wild-type with respect to the growth of Psm ES4326 carrying avrRpt2. Furthermore, leaves of the four mutant plants infected with a high dose (10⁶ cfu/cm² leaf area) of Psm ES4326/avrRpt2 gave rise to a macroscopically visible HR, indistinguishable from the response in wild type, manifested by characteristic tissue collapse within 20 hours after infection (data not shown). We therefore conclude that the four eds mutants maintain full resistance to the bacterial avirulence gene avrRpt2, including the capacity to mount an HR.

View larger version (34K): In this window In a new window Download PPT slide   Figure 2. —Growth of virulent and isogenic avirulent strains of Psm ES4326 in wild-type Arabidopsis (Columbia ecotype carrying BGL2-GUS transgene) leaves and eds mutant leaves. Leaves were inoculated with Psm ES4326 or Psm ES4326 carrying avrRpt2 at a dose of 103 cfu/cm2. Three days postinoculation, infected leaves were excised and assayed for bacterial density as described in MATERIALS AND METHODS. The bars represent the mean and standard deviation of values from four replicate samples. This experiment was repeated with similar results. The data for mutant npr1-4 were obtained from different experiments and normalized to the values for wild-type plants.

Avirulent pathogen-mediated SAR response in eds mutants:
We have demonstrated that the mutants npr1-4, eds5-2, eds10-1, eds11-1, eds12-1, and eds13-1 retain the capacity to recognize bacterial pathogens carrying the avirulence gene avrRpt2 and to suppress their growth in infected leaves. In Arabidopsis wild-type plants, the recognition of a bacterial avirulence gene by the corresponding resistance gene initiates a signal cascade that results in the activation of a plant defense mechanism known as systemic acquired resistance (SAR) (ENYEDI et al. 1992B ; RYALS et al. 1994 ). Characteristics of SAR include the accumulation of salicylic acid, the increased expression of pathogenesis-related (PR) genes, and the induction of systemic resistance to a broad spectrum of pathogens (WARD et al. 1991 ; GAFFNEY et al. 1993 ; UKNES et al. 1993 ; DELANEY et al. 1994 ). One class of eds mutants, npr1, is not only debilitated in restricting the growth of Psm ES4326 in infected leaves but is also compromised in mounting SAR in response to a variety of known inducers (CAO et al. 1994 ). Thus, we tested whether the eds mutants npr1-4, eds5-2, eds10-1, eds11-1, eds12-1, and eds13-1 are affected in their ability to induce SAR upon challenge with an avirulent pathogen (essentially as described in ROGERS and AUSUBEL 1997 ). The bean pathogen Pseudomonas syringae pv. phaseolicola 3121 carrying avrRpt2 was chosen to trigger the SAR response. This strain has been shown to induce a strong HR in Arabidopsis carrying the resistance gene RPS2 (YU et al. 1993 ) and can be easily distinguished from the challenging virulent strain Psm ES4326 based on antibiotic resistance.

Three fully expanded leaves per plant of npr1-4, eds5-2, eds10-1, eds11-1, eds12-1, eds13-1, and Col-0 wild type were inoculated with either P. s. phaseolicola 3121/avrRpt2 at the dose of 10⁶ cfu/cm² or with 10 mM MgSO₄ as a control (mock treatment). Eight days later, two uninfected leaves on the same plants were inoculated with Psm ES4326 at the dose of 10³ cfu/cm², and the density of the virulent pathogen was determined. As shown in Figure 3, the growth of Psm ES4326 was at least 10-fold reduced in wild-type plants and the five mutants eds5-2, eds10-1, eds11-1, eds12-1, and eds13-1 pretreated with P. s. phaseolicola 3121, compared with the growth in plants that were mock-treated with 10 mM MgSO₄. However, as shown previously (CAO et al. 1994 ) this difference in Psm ES4326 growth was not observed in npr1-4 plants, which are blocked in the SAR response downstream of SA perception. These results clearly show that the five eds mutants (eds5-2, eds10-1, eds11-1, eds12-1, and eds13-1) retain the capacity to initiate at least a partial SAR response following treatment with an avirulent pathogen. Interestingly, even after the induction of SAR, the five eds mutant plants displayed enhanced growth of Psm ES4326 in comparison to wild-type Columbia.

View larger version (56K): In this window In a new window Download PPT slide   Figure 3. —Induction of SAR in Arabidopsis wild-type (Columbia ecotype carrying the BGL2-GUS transgene) leaves and eds mutant leaves. Eight days prior to the secondary challenge with Psm ES4326 at a dose of 103 cfu/cm2, three lower leaves per plant were inoculated with either 10 mM MgSO4 as a control () or with 106 cfu/cm2 P. s. phaseolicola 3121 carrying avrRpt2 to induce SAR (). Three days after the secondary infection with Psm ES4326, infected leaves were excised and bacterial density was determined as log (cfu/cm2 leaf area). Each bar represents the mean and standard deviation of six replicate samples. The experiment was repeated with similar results.

Phenotypes of eds mutants in response to different bacterial phytopathogens:
The six eds mutants npr1-4, eds5-2, eds10-1, eds11-1, eds12-1, and eds13-1 were isolated on the basis of their enhanced disease susceptibility phenotypes in response to Psm ES4326 infection. To test whether the mutants are specifically affected in restricting the growth of Psm ES4326 or whether these EDS genes are also involved in regulating the growth of a variety of virulent Col-0 bacterial pathogens, we extended the characterization of these eds mutants with respect to their response to the phytopathogenic strain P. syringae pv. tomato (Pst) DC3000, the multi-host pathogen P. aeruginosa UCBPP-PA14 (PA14), and the leaf-spotting pathogen Xanthomonas campestris pv. raphani (Xcr) 1946.

Wild-type, npr1-4, eds5-2, eds10-1, eds11-1, eds12-1, and eds13-1 mutant plants were inoculated with these bacterial pathogens at the doses indicated (Figure 4, A–C), and the bacterial growth was assayed three days after inoculation. As shown in Figure 4A, the final titer of Pst DC3000 was at least 10-fold higher in all six eds mutants than it was in wild-type plants, indicating that these mutations also cause enhanced susceptibility to Pst DC3000. Furthermore, the symptoms of the eds mutants in response to Pst DC3000 infection were indistinguishable from those displayed after Psm ES4326 infection. These results are consistent with previous studies indicating that these two P. syringae pathogens are very similar in terms of the severity of disease symptoms they elicit in Col-0 leaves as well as of other disease-related characteristics (DAVIS et al. 1991 ; DONG et al. 1991 ; WHALEN et al. 1991 ).

View larger version (40K): In this window In a new window Download PPT slide   Figure 4. —Growth of bacterial pathogens in wild-type Arabidopsis (Columbia ecotype carrying BGL2-GUS transgene) leaves and eds mutant leaves. (A) Growth of P. s. tomato (Pst) DC3000. Three days after inoculation with Pst DC3000 at a density of 103 cfu/cm2, leaves were excised and the bacterial density was determined. Each bar represents the mean and standard deviation of values from four replicate samples. This experiment was repeated twice with similar results. (B) Growth of P. aeruginosa UCBPP-PA14 (PA14). Three days after inoculation with PA14 at a density of 103 cfu/cm2, leaves were excised and the bacterial density was determined. Each bar represents the mean and standard deviation of values from four replicate samples. This experiment was repeated with similar results. (C) Growth of X. c. raphani (Xcr) 1946. Three days after inoculation with Xcr 1946 at a density of 104 cfu/cm2, leaves were excised and the bacterial density was determined. Each bar represents the mean and standard deviation of values from four replicate samples. This experiment was repeated twice with similar results. In all three panels, the data for mutant npr1-4 were obtained from different experiments and were normalized to values obtained for wild-type plants.

Interestingly, Figure 4B shows that all six eds mutations are also more susceptible to P. aeruginosa UCBPP-PA14, a clinical isolate of the opportunistic pathogen that has been reported to be pathogenic on Arabidopsis in an ecotype-specific manner (RAHME et al. 1995 ; RAHME et al. 1997 ). Enhanced growth of P. aeruginosa in the eds mutants correlated with enhanced symptom formation. On the other hand, Figure 4C shows that only three of the six eds mutants (eds5-2, eds10-1, and eds13-1) are slightly more susceptible to Xanthomonas campestris pv. raphani 1946, a leaf-spotting pathogen that is virulent on Col-0 wild-type plants. Consistent with these growth results, only very mild, if any, disease symptoms were observed in eds mutants infected with X. campestris. These results indicate that enhanced disease symptom development correlates with enhanced bacterial growth in these six eds mutants.

Some EDS genes play a role in restricting the growth of the fungal pathogen Erysiphe orontii:
It was of interest for us to determine whether the eds mutants are affected specifically in their susceptibility to bacterial pathogens, or whether the corresponding EDS genes are also involved in restricting the growth of eukaryotic pathogens, such as fungi. Erysiphe sp. are obligate biotrophic parasites that cause powdery mildew disease on a variety of plant species, including the Brassicaceae (CRUTE et al. 1994 ). The disease phenotype starts with the development of white fungal colonies on the surface of infected leaves resulting in the formation of a white mat of mycelia and conidiophores, visible by eye at seven to ten days postinoculation. For our studies we used an isolate of E. orontii MGH, which was recovered from Arabidopsis plants grown in our greenhouse at the Massachusetts General Hospital (J. PLOTNIKOVA, T. REUBER, F. AUSUBEL and D. PFISTER, unpublished results). We assayed the extent of E. orontii MGH growth on infected leaves of wild-type and mutant plants at 7 and 16 dpi, using IR scores based on visual examination of mycelial growth (see MATERIALS AND METHODS for a complete description of the scoring method). Wild-type and eds mutant plants were grown side by side in the same pot and fungal conidia were applied to the leaf surfaces using a settling tower. At 7 dpi, wild-type plants, npr1-4, eds5-2, eds11-1, and eds12-1 did not show any symptoms of powdery mildew disease, whereas on leaves of mutants eds10-1 and eds13-1 singular white spots of powdery mildew were readily visible (Table 6). When examined at 16 dpi, infected leaves of npr1-4, eds5-2, eds10-1, and eds13-1 displayed clearly enhanced fungal growth in comparison to wild type, with approximately 50% of the infected eds leaf area covered by a white mycelial mat. At this time point, the powdery mildew symptoms displayed on eds12-1 could not be distinguished visually from that on wild-type plants, and thus were given the same IR score. Interestingly, the growth of the fungal pathogen on eds11-1 was reduced compared to the growth on wild-type plants at 16 dpi. However, the pathogen continued to develop slowly on eds11-1 with longer incubation periods, suggesting that the growth was delayed, but not inhibited.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

To identify previously unknown genes that play a role in restricting pathogen growth, we performed a direct screen for Arabidopsis mutants that display an eds phenotype. We obtained 11 putative eds mutants which, when compared to wild type, supported at least 10-fold higher growth of the virulent pathogen P. syringae pv. maculicola (Psm) ES4326 following a low-dose inoculation. These 11 eds mutants were not affected in the synthesis of camalexin in response to Psm ES4326 and also were not affected in their ability to mount an HR in response to the isogenic avirulent strain carrying the avirulence gene avrRpt2.

The eds mutants were isolated in a transgenic plant carrying a chimeric BGL2-GUS construct which served as a convenient reporter for pathogen-induced activation of the PR gene BGL2. The 11 eds mutants were tested for the activation of the BGL2-GUS transgene in response to treatment with SA. Ten of the eds mutants showed SA-mediated activation of the transgene indicating that these mutants do not fall into the class of the previously characterized npr1/nim1/sai1 mutant (CAO et al. 1994 ; DELANEY et al. 1995 ; SHAH et al. 1997 ). eds-17, the only mutant that failed to induce the transgene BGL2-GUS in response to SA, was found to carry a mutant allele of NPR1 determined by complementation analysis and was renamed npr1-4. Furthermore, this genetic result was confirmed by direct DNA sequence analysis of the eds-17 npr1 gene (CAO et al. 1997 ). Although the other eds mutants exhibited significant SA-induced BGL2-GUS activity, identically treated eds-215 (eds13-1) plants showed levels of GUS activity at about 25% of wild type. Partial activation of a PR gene by SA is a phenotype that has not previously been reported for a defense-related gene. It does not seem likely that eds-215 (eds13-1) is a leaky npr1 mutant since eds-215 (eds13-1) fully complemented the eds phenotype of npr1-4. We are currently in the process of measuring both the accumulation of endogenous PR transcripts and endogenous SA levels in eds-215 (eds13-1) in response to SA treatment.

The genetic analysis presented in Table 5 shows that all eds mutants as well as npr1-4 segregate as single recessive Mendelian genes. Four of the eds mutants, eds10-1 (eds-15), eds11-1 (eds-16), eds12-1 (eds-111), and eds13-1 (eds-215), define new genetic loci based on complementation testing with eds mutants isolated in previous screens (CAO et al. 1997 ; GLAZEBROOK et al. 1996 ; PARKER et al. 1996 ; ROGERS and AUSUBEL 1997 ). The mutant eds-19 was found to be an allele of the previously described EDS5 gene (GLAZEBROOK et al. 1996 ; ROGERS and AUSUBEL 1997 ) and was renamed eds5-2. To date, a total of 14 eds mutants isolated in our laboratory have been subjected to complementation analysis. These mutations define 12 EDS genes, indicating that the screen for EDS genes is not nearly saturated and that many factors involved in defense responses remain to be discovered.

The overall phenotype of eds5-2 (isolated in this study) in response to challenge with different bacterial pathogens is similar to that reported previously for eds5-1 (ROGERS and AUSUBEL 1997 ). The npr1-4 mutant (isolated in this study) also has the same overall phenotype as previously isolated npr1/nim1/sai1 mutants (CAO et al. 1994 ; DELANEY et al. 1995 ; GLAZEBROOK et al. 1996 ; SHAH et al. 1997 ). These latter mutants block the induction of the PR genes PR1, BGL-2 (PR2) and PR5 in response to SA, fail to mount SAR in response to a variety of known inducers and display enhanced susceptibility to virulent P. syringae strains. However, whereas npr1-4 supported 50–100-fold enhanced growth of the virulent pathogen Psm ES4326, no enhanced growth was observed following infection with the isogenic avirulent strain under the conditions used in our laboratory. Other studies involving different npr1 alleles and different avirulent pathogens indicate that NPR1 plays a role in restricting the growth of avirulent pathogens (DELANEY et al. 1995 ; SHAH et al. 1997 ). The sai1 mutant (SHAH et al. 1997 ) for example, another allele of npr1 that displays phenotypes similar to the npr1 and nim1 mutants, exhibited enhanced growth of avirulent P. syringae strains. One explanation for the seemingly contradictory findings concerning npr1-4 and other npr1 alleles is that npr1-4 is leaky. However, no BGL2-GUS expression in response to SA was observed and the induction of SAR following pretreatment with a necrotizing avirulent pathogen was completely blocked in this mutant (see Figure 3). Our results are consistent with the conclusion that the defense response pathways activated downstream of NPR1 do not play a significant role in restricting the growth of avirulent bacterial pathogens. Clarification of this discrepancy awaits side-by-side comparison of all npr1 alleles challenged with the same pathogen under identical conditions.

Similar to the results we obtained for npr1-4, the remaining five eds mutants (eds10-1, eds11-1, eds12-1, eds13-1, and eds5-2) did not support increased growth of the avirulent strain Psm ES4326/avrRpt2. These findings indicate that all of the eds mutants, including npr1-4, retain the ability to recognize the product of the avirulence gene avrRpt2 and that the defense response pathways activated by avr-R gene interaction are sufficient to limit the growth of avirulent pathogens in the absence of defense functions provided by EDS genes.

As reported previously for several other eds mutants (ROGERS and AUSUBEL 1997 ), the data reported in Figure 3 show that the signal transduction pathway(s) leading to the induction of SAR appear to be at least partially intact in eds10-1, eds11-1, eds12-1, eds13-1, and eds5-2. Moreover, consistent with data obtained previously for other eds mutants, the eds mutants isolated in this study allowed enhanced growth of Psm ES4326 even after the induction of SAR in comparison to wild-type plants. This indicates that the signal transduction pathways leading to the activation of local defense responses and SAR may share common factors. The eds phenotypes observed do not rule out the possibility that the eds mutations lead to a reduction in endogenous SA levels. However, at least in the case of npr1/nim1/sai1 it has been shown that the SAR pathway is blocked downstream of SA accumulation.

The five eds mutants and npr1-4 characterized here affect diverse defense-related functions resulting in distinguishable phenotypes. All of these mutants display enhanced susceptibility to the closely related pseudomonads P. s. maculicola ES4326 and P. s. tomato DC3000, as well as to the distantly related P. aeruginosa UCBPP-PA14. However, only a subset of these mutants, eds5-2, eds10-1, and eds13-1, also display enhanced susceptibility to the leaf-spotting pathogen X. c. raphani 1946. Similarly, npr1-4, eds10-1, eds13-1, and eds5-2 are more susceptible to the obligate biotrophic fungus Erysiphe orontii whereas the phenotypes of eds11-1 and eds12-1 are indistinguishable from wild type after infection with the fungus. The data in Table 6 show that the defense responses defined by EDS genes are not equally effective against different pathogens and suggest that screens to isolate eds mutants using different pathogens will lead to the identification of novel EDS genes. Such a screen is currently being carried out in our laboratory using Erysiphe orontii MGH as the challenging pathogen (T. L. REUBER, unpublished data).

The fact that the genetic characterization presented in this and other studies revealed only a small number of allelic pairs among the eds mutants isolated to date indicates that there are a large number of nonredundant defense responses or defense response signaling pathways or both involved in limiting the growth of Psm ES4326. There is increasing evidence that plant defense responses can act in concert to restrict pathogen growth. For example, the PR proteins chitinase and glucanase have been shown to act synergistically to restrict the growth of certain fungal pathogens in vitro or when overexpressed in transgenic tobacco (MAUCH et al. 1988 ; ZHU et al. 1994 ). If these synergistic interactions are common, a mutation in any of these defense responses may cause an enhanced susceptibility phenotype and might explain why mutations in so many defense-related genes have such strong phenotypes.

The eds mutants presented in this study are not equally susceptible to different prokaryotic and eukaryotic pathogens. These observations indicate that the defense responses affected in these eds mutants are not equally effective in restricting pathogen growth. Mutants, such as npr1-4, eds5-2, eds10-1, and eds13-1, that exhibit an eds phenotype in response to a variety of pathogens are likely to carry mutations in regulatory factors rather than in downstream components of pathogen-specific defense mechanisms. Mutations in the PAD4 gene, for example, render plants highly susceptible to both P. syringae strains as well as various phytopathogenic fungi, making this gene a candidate for encoding a pleiotropic regulator of defense responses (GLAZEBROOK et al. 1997B ; T. L. REUBER, personal communication).

The data presented in this paper provide further support for the conclusion that direct screening for mutants with increased pathogen-susceptibility is a useful strategy for identifying Arabidopsis genes that encode defense-related functions. Cloning and further analysis of EDS genes should lead to new insights into the mechanisms effective in combating pathogen attack.

	ACKNOWLEDGMENTS

We thank XINNIAN DONG for providing the EMS-mutagenized BGL2-GUS seeds, JACQUELINE HEARD, T. LYNNE REUBER, JULIE STONE, PETE YORGEY, and other colleagues in the laboratory for many helpful discussions and for advice on the preparation of the manuscript. This work was supported by National Institutes of Health grant GM-48707 and by a grant from Monsanto Company awarded to F.M.A.

Manuscript received January 21, 1998; Accepted for publication March 4, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ADAM, L. and S. C. SOMERVILLE, 1996  Genetic characterization of five powdery mildew disease resistance loci in Arabidopsis thaliana.. Plant J. 9:341-356[Medline].

ALEXANDER, D., R. M. GOODMAN, M. GUT-RELLA, C. GLASCOCK, and K. WEYMANN et al., 1993  Increased tolerance to two oomycete pathogens in transgenic tobacco expressing pathogenesis-related protein 1a. Proc. Natl. Acad. Sci. USA 90:7327-7331[Abstract/Free Full Text].

BROGLIE, K., I. CHET, M. HOLLIDAY, R. CRESSMAN, and P. BIDDLE et al., 1991  Transgenic plants with enhanced resistance to the fungal pathogen Rhizoctonia solani.. Science 254:1194-1197[Abstract/Free Full Text].

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

CAO, H., J. GLAZEBROOK, J. D. CLARKE, S. VOLKO, and X. DONG, 1997  The Arabidopsis NPR1 gene that controls systemic acquired resistance encodes a novel protein containing ankyrin repeats. Cell 88:57-63[Medline].

CRUTE, I. R. and D. A. C. PINK, 1996  Genetics and utilisation of pathogen resistance in plants. Plant Cell 8:1747-1755[Medline].

CRUTE, I., J. BEYNON, J. DANGL, E. HOLUB, B. MAUCH-MANI et al., 1994 Microbial pathogenesis of Arabidopsis, pp. 705–747 in Arabidopsis, edited by E. M. MEYEROWITZ, and C. R. SOMERVILLE. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.

CUPPELS, D. A., 1986  Generation and characterization of Tn5 insertion mutations in Pseudomonas syringae pv. tomato. Appl. Environ. Microbiol. 52:323-327.

DANGL, J. L., 1993  Applications of Arabidopsis thaliana to outstanding issues in plant-pathogen interactions. Int. Rev. Cytol. 144:53-83.

DAVIS, K. R., E. SCHOTT, and F. M. AUSUBEL, 1991  Virulence of selected phytopathogenic pseudomonads in Arabidopsis thaliana.. Mol. Plant-Microbe Interact. 4:477-488.

DELANEY, T. P., S. UKNES, B. VERNOOIJ, L. FRIEDRICH, and K. WEYMANN et al., 1994  A central role of salicylic acid in plant disease resistance. Science 266:1247-1250[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].

DIXON, R. A. and C. J. LAMB, 1990  Molecular communication in interactions between plants and microbial pathogens. Annu. Rev. Plant Physiol. Plant Mol. Biol. 41:339-367.

DIXON, R. A., M. J. HARRISON, and C. J. LAMB, 1994  Early events in the activation of plant defense responses. Annu. Rev. Phytopathol. 32:479-501.

DONG, X., M. MINDRINOS, K. R. DAVIS, and F. M. AUSUBEL, 1991  Induction of Arabidopsis defense genes by virulent and avirulent Pseudomonas syringae strains and by a cloned avirulence gene. Plant Cell 3:61-72[Abstract/Free Full Text].

ENYEDI, A. J., N. YALPANI, P. SILVERMAN, and I. RASKIN, 1992a  Signal molecules in systemic plant resistance to pathogens and pests. Cell 70:879-886[Medline].

ENYEDI, A. J., N. YALPANI, P. SILVERMAN, and I. RASKIN, 1992b  Localization, conjugation, and function of salicylic acid in tobacco during the hypersensitive reaction to tobacco mosaic virus. Proc. Natl. Acad. Sci. USA 89:2480-2484[Abstract/Free Full Text].

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

GAFFNEY, T., L. FRIEDRICH, B. VERNOOIJ, D. NEGROTTO, and G. NYE et al., 1993  Requirement of salicylic acid for the induction of systemic acquired resistance. Science 261:754-756.

GLAZEBROOK, J. 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., E. E. ROGERS, and F. M. AUSUBEL, 1997a  Use of Arabidopsis for genetic dissection of plant defense responses. Annu. Rev. Genet. 31:547-569[Medline].

GLAZEBROOK, J., M. ZOOK, E. B. HOLUB, I. KAGAN, and E. E. ROGERS et al., 1997b  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].

HAIN, R., H.-J. REIF, E. KRAUSE, R. LANGEBARTELS, and H. KINDL et al., 1993  Disease resistance results from foreign phytoalexin expression in a novel plant. Nature 361:153-156[Medline].

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

JACH, G., B. GORNHARDT, J. MUNDY, J. LOGEMANN, and E. PINSDORF et al., 1995  Enhanced quantitative resistance against fungal disease by combinatorial expression of different barley antifungal proteins in transgenic tobacco. Plant J. 8:97-109[Medline].

KAUFFMANN, S., M. LEGRAND, P. GEOFFERY, and B. FRITIG, 1987  Biological function of `pathogenesis-related' proteins: four PR proteins of tobacco have 1,3-ß-glucanase activity. EMBO J. 6:3209-3212[Medline].

KUNKEL, B. N., 1996  A useful weed put to work: genetic analysis of disease resistance in Arabidopsis thaliana.. Trends Genet. 12:62-69.

LAMB, C. J., 1994  Plant disease resistance genes in signal perception and transduction. Cell 76:419-422[Medline].

LAMB, C. J., M. A. LAWTON, M. DRON, and R. A. DIXON, 1989  Signals and transduction mechanisms for activation of plant defenses against microbial attack. Cell 56:215-224[Medline].

LEGRAND, M., S. KAUFFMANN, P. GEOFFREY, and B. FRITIG, 1987  Biological function of pathogenesis-related proteins: four tobacco pathogenesis-related proteins are chitinases. Proc. Natl. Acad. Sci. USA 84:6750-6754[Abstract/Free Full Text].

LIU, D., K. G. RAGHOTHAMA, P. M. HASEGAWA, and R. A. BRESSAN, 1994  Osmotin overexpression in potato delays development of disease symptoms. Proc. Natl. Acad. Sci. USA 91:1888-1892[Abstract/Free Full Text].

MALAMY, J. and D. F. KLESSIG, 1992  Salicylic acid and plant disease resistance. Plant J. 2:643-654.

MAUCH, F., L. A. HADWIGER, and T. BOLLER, 1988  Antifungal hydrolases in pea tissue. I. Plant Physiol. 87:325-333[Abstract/Free Full Text].

MELCHERS, L. S., M. A.-D. GROOT, J. A. V. D. KNAPP, A. S. PONSTEIN, and M. B. SELA-BUURLAGE et al., 1994  A new class of tobacco chitinases homologous to bacterial exo-chitinases displays anti-fungal activity. Plant J. 5:469-480[Medline].

NIDERMAN, T., I. GENETET, T. BRUYERE, R. GEES, and A. STINTZI et al., 1995  Pathogenesis-related PR-1 proteins are antifungal. Plant Physiol. 108:17-27[Abstract].

PARKER, J. E., C. E. BARBER, F. MI-JIAO, and M. J. DANIELS, 1993  Interaction of Xanthomonas campestris with Arabidopsis thaliana: characterization of a gene from X. campestris pv. raphani that confers avirulence on most A. thaliana accessions. Mol. Plant-Microbe Interact. 6:216-224[Medline].

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

PONSTEIN, A. S., S. A. BRES-VLOEMANS, M. B. SELA-BUURLAGE, P. J. M. VAN DEN ELZEN, and L. S. MELCHERS et al., 1994  A novel pathogen- and wound-inducible tobacco (Nicotiana tabacum) protein with antifungal activity. Plant Physiol. 104:109-118[Abstract].

RAHME, L. G., M. N. MINDRINOS, and N. J. PANOPOULOS, 1992  Plant and environmental sensory signals control the expression of hrp genes in Pseudomonas syringae pv. phaseolicola. J. Bacteriol. 174:3499-3507[Abstract/Free Full Text].

RAHME, L. G., E. J