Jasmonate signaling mutants ofArabidopsisare susceptible to the soil fungusPythium irregulare

Authors


*For correspondence (fax +402 472 7904;
e-mail pstaswick@crcvms.unl.edu).

Summary

Jasmonic acid has properties of a plant hormone, including the induction of specific genes associated with plant defense. We previously describedjar1-1, anArabidopsisjasmonate response mutant that exhibits reduced sensitivity to methyl jasmonate. We have further characterized this mutant and two new alleles;jar1-2from a gamma irradiated population, andjar1-4from a T-DNA mutant population. Seedling root growth injar1-1was equally insensitive to methyl jasmonate and jasmonic acid, indicating that the defect was not in the conversion of methyl jasmonate to the acid. None of thejar1mutants showed an altered sensitivity to auxin, cytokinin, or the ethylene precursor 1-aminocyclopropane-1-carboxylic acid, indicating that the lesion does not affect the general uptake or transport of hormones. A soil fungus,Pythium irregulare, was found to blightjar1-1. Cultures of this organism caused the symptoms in all threejar1mutants but not in wild type, indicating that increased susceptibility was due to the lesion in theJAR1locus. A fatty acid desaturase triple mutant that is defective in the biosynthesis of jasmonic acid (J. Browse, Washington State University) was also susceptible, confirming that jasmonate is involved in resistance. Thejar1-1locus was mapped to the lower end of chromosome 2, about 11.4 cM fromas1 and1.6 cM fromcer8. These results establish that jasmonate signaling plays an important role in resistance to soil micro-organisms in plants.

Introduction

Plants utilize a variety of strategies, both passive and inducible, to resist infection by pathogenic micro-organisms (reviewed in Hammond-Kosack & Jones 1996). Signaling molecules implicated in inducible defense systems include plant hormones, salicylic acid ( Delaney et al. 1994 ;Ryals et al. 1995 ), H2O2 ( Levine et al. 1994 ;Mehdy 1994) and various ions ( Ward et al. 1995 ). Recent evidence indicates that jasmonic acid (JA) also plays an important role in plant defense signaling pathways.

Jasmonate stimulates the expression of several genes associated with plant stress response. The most intensely studied genes that are activated by jasmonate are the wound inducible proteinase inhibitors of Solonaceae plants that can protect against insect feeding ( Farmer & Ryan 1990, 1992). Others include the vegetative storage protein genes of soybean ( Staswick 1994) and genes associated with defense against micro-organisms, such as the production of phytoalexins ( Mueller et al. 1993 ).

Mutants that are defective in their response to specific plant hormones have proven to be powerful tools for the analysis of hormone function, the molecular genetic dissection of signaling pathways, and for the isolation of the corresponding genes. We described the first plant jasmonate signaling mutant which was isolated in a screen for exceptional Arabidopsis seedlings that were less inhibited in root elongation than wild-type when grown in the presence of methyl jasmonate (MeJA) ( Staswick et al. 1992 ). In addition to the signaling defect related to root growth, the lesion in jar1-1 also affected the MeJA-induced accumulation of vegetative storage protein (VSP) homologues in leaves. Two additional Arabidopsis jasmonate response loci were subsequently identified, jin1 ( Berger et al. 1996 ) and coi1 ( Feys et al. 1994 ;Xi et al. 1998 ).

Another class of jasmonate mutant is defective in jasmonate biosynthesis. A fatty acid desaturase (fad) triple mutant is essentially devoid of jasmonate because it does not produce linolenic acid, the initial substrate for jasmonate biosynthesis ( McConn & Browse 1996). Among all of the available jasmonate mutants, the only observed abnormality in growth and development was male sterility in the coi1 and fad mutants. Tomato mutants (e.g. def1) defective in octadecanoid metabolism also develop normally ( Howe et al. 1996 ;Lightner et al. 1993 ). The accumulating evidence suggests that the primary function of jasmonate is not in growth and development but in plant stress or defensive responses.

Evidence that jasmonate signaling is involved in protection against insects in plants is well established. Proteinase inhibitor genes are activated by exogenously applied jasmonate ( Farmer & Ryan 1992), and jasmonate levels rise in response to a wound stimulus that at least partially mimics the response to insect feeding ( Green & Ryan 1972;Korth & Dixon 1997). More directly, mutants affecting jasmonate accumulation exhibit decreased resistance to insect attack. The def1 mutant of tomato is more susceptible to tobacco hornworm larvae ( Howe et al. 1996 ) and the Arabidopsis fad triple mutant is much more susceptible to fungal gnats than are wild-type plants ( McConn et al. 1997 ).

Evidence that octadecanoid signaling is directly involved in the defense against micro-organisms has been less direct. Exogenous jasmonate induces genes associated with phytoalexin biosynthesis and leads to the accumulation of these antimicrobial compounds ( Choi et al. 1994 ;Mueller et al. 1993 ). Fungal elicitors of phytoalexin biosynthesis also stimulate the accumulation of jasmonic acid in suspension cell cultures ( Gundlach et al. 1992 ). However, to our knowledge evidence that mutants in octadecanoid signaling are more susceptible to micro-organisms has not been reported. We document here that Arabidopsis mutants defective in jasmonate signaling are much more susceptible than wild-type to the soil fungus Pythium irregulare. These results provide a new opportunity to understand the function of jasmonate in plants, as well as the mechanisms for resistance to Pythium, an economically important genus of pathogens.

Results

The jar1 mutant is less sensitive to both jasmonic acid and methyl jasmonate

Both JA and MeJA are present in plants and both can have biological activity when applied exogenously, although sensitivity to each reportedly varies for some responses. Since the initial screening and subsequent characterization of jar1-1 utilized MeJA, it could be argued that resistance was due to the inability of this mutant to demethylate MeJA to the active acid. To test this, jar1-1 and wild-type seedlings were grown on agar plates oriented vertically and containing either JA or MeJA. Mean root length was determined for 25 seedlings after 10 days. Lengths (mm ± SD) on 10 μm MeJA or JA were 9.8 ± 1.2 and 9.6 ± 1.1, respectively, for wild type; and 18.4 ± 2.6 mm and 18.8 ± 1.9 mm, respectively, for jar1-1. Similarly, root growth of jar1-1 was approximately twice that of wild type on 100 μm MeJA or JA, and there was no significant difference between the results for the acid and the methylester. Previous studies established that there was no difference in root growth in the absence of jasmonate ( Staswick et al. 1992 ). These results demonstrate that the insensitivity in jar1-1 is not due to a defect in converting MeJA to the acid.

Insensitivity of jar1 is specific for jasmonate

To examine the specificity of the jar1 locus for jasmonate signaling, we tested the effect of auxin, ethylene and cytokinin in the root growth assay on agar plates. The ethylene precursor 1-aminocyclopropane-1-carboxylic acid (ACC) was used in this experiment because it was easily incorporated into the agar medium. Two additional mutants were evaluated along with jar1-1. Both were found to be allelic to jar1-1 by genetic complementation (data not shown). The results shown in Table 1 indicate that mutation of the JAR1 locus does not confer altered sensitivity to these growth regulators. Thus, the locus appears to be quite specific for insensitivity to jasmonate signaling.

Table 1.  Inhibition of jar1 seedling root growth by various plant hormones
Hormone (μm) Wild typejar1-1jar1-2jar1-4
  1. Values are the mean root length (mm) of 10 seedlings ± SD measured 7 days after sowing.

  2. 1n1">1BA = N6-benzyladenine; IAA = Indole acetic acid, ACC = 1-aminocyclopropane-1-carboxylic acid.

None10.2 ± 0.79.7 ± 0.410.0 ± 0.69.7 ± 0.6
BA 1 (0.5) 4.7 ± 0.24.4 ± 0.2 4.6 ± 0.24.6 ± 0.2
IAA (1.0) 1.3 ± 0.11.3 ± 0.2 1.2 ± 0.21.3 ± 0.1
ACC (1.0) 2.0 ± 0.52.1 ± 0.3 2.0 ± 0.01.9 ± 0.3

Jasmonate responsive genes are affected differentially in jar1-1

The induction of a jasmonate responsive VSP homologue gene in leaves was investigated by monitoring steady state mRNA levels in mutant and jar1-1 plants. Figure 1 shows that the uninduced level of VSP mRNA was higher in wild type than in the mutant. Accumulation of VSP mRNA in response to MeJA treatment occurred in both mutant and non-mutant plants, although the overall accumulation was greater for wild type than for jar1-1. Linolenic acid, a precursor to jasmonate, also stimulated mRNA accumulation in both the wild type and the mutant, and again the overall accumulation was higher in the wild type.

Figure 1.

Induction of VSP-like mRNA in leaves by arachidonate and linolenate.

Fifteen-day-old jar1-1 (J) or wild-type (WT) plants were treated with 20 μm methyl jasmonate (MeJA), or 50 μm arachidonic (Arach), or linolenic acid (Linol) dissolved in ethanol (0.001% (v/v) final concentration). Controls were untreated at time 0 or treated 48 h with ethanol (EtOH) alone.

The jar1-1 lesion affects some but not all responses to arachidonic acid

The elicitor arachidonate derived from plant fungal pathogens can signal expression of genes presumed to be involved in plant defense responses ( Choi et al. 1994 ) and arachidonate is found in several plant pathogens, including some Pythium species ( Weete & Weber 1980). Therefore, we investigated whether the defect in jar1-1 affected the response to arachidonic acid. Figure 1 shows that arachidonate also induced VSP homologue mRNA accumulation in the leaves of both wild-type and mutant plants. However, the response was weaker in jar1-1 plants and the magnitude of the reduction in response to arachidonate in jar1-1 was comparable to that found for MeJA and linolenic acid. In contrast to this result, Table 2 shows that there was no difference between mutant and wild type in the level of inhibition of root growth caused by arachidonate.

Table 2.  Sensitivity of jar1-1 and wild-type root growth to linolenate and arachidonate
TreatmentWild typejar1-1
  1. Values are the mean root length (mm) ± SD of 20 seedlings 8 days after sowing. The concentration of arachidonic and linolenic acid were 50 μm and MeJA was 20 μm.

None29.6 ± 1.529.9 ± 1.6
Arachidonate19.7 ± 1.419.9 ± 0.6
Linolenate21.1 ± 1.521.4 ± 1.7
MeJA 9.9 ± 0.920.2 ± 1.3

An isolate of Pythium irregulare preferentially infects jar1 mutants

We noted that occasionally jar1-1, but not wild type, exhibited loss of turgor followed by tissue collapse and plant death. The radial pattern of symptom spread over a period of several days suggested that a soil pathogen was involved. This hypothesis was supported by the fact that soil from affected pots was able to reproduce the symptoms when applied to the mutant, but not in wild-type plants (data not shown). We isolated the fungus Pythium irregulare from affected plants on two different occasions, and the fungus was not detected in asymptomatic plants. Figure 2 shows that inoculation of the root zone with pure cultures of this fungal isolate caused the same symptoms observed earlier on jar1-1. Within 2–3 days, leaf coloration changed from a bright light green to a duller appearance. By 3–4 days plants began showing signs of wilting followed by collapse of leaves and subsequent death of tissue. Mutant and wild-type plants were indistinguishable prior to inoculation (not shown).

Figure 2.

Infection of jar1-1 and fad triple mutants by P. irregulare.

Plant clusters were inoculated 4 weeks after planting and the photograph was taken 10 days later. Mock inoculated controls showed no symptoms (not shown).

In order to establish that susceptibility in jar1-1 was not the result of a mutation in another gene, we tested two additional alleles of the jar1 locus. Figure 3 shows that for all three alleles all seedlings were affected within 1 week after inoculation. Mock inoculated plants never showed symptoms (not shown). In contrast, only rarely did wild-type plants show symptoms even after 10 days. These results demonstrate that a functional JAR1 locus is necessary for resistance to this isolate of P. irregulare in Arabidopsis.

Figure 3.

Time course of infection for three different alleles of jar1 by P. irregulare.

The experimental unit consisted of 21-day-old clusters of about 20 seedlings each of wild-type and of a mutant grown in a single pot. Pots were counted as showing symptoms when the first signs of leaf wilting were observed. Values shown are the mean of two independent experiments and error bars indicate the standard deviation. N = 15 pots/experiment for each data point. The wild-type data shown is for the Columbia ecotype. The wild-type control for jar1-4 was the ecotype Wassilewskija, which showed no symptoms throughout the experiment.

Although the temporal development of jar1-1 plants is visibly indistinguishable from wild type, we investigated whether susceptibility of the mutant or wild type was affected by plant age. Plants were grown in soil for 16, 22, 31 or 39 days and then inoculated at the same time from a common culture of P. irregulare. Figure 4 shows that while there was a tendency for younger plants to display symptoms earlier than older plants, even 39-days-old jar1-1 plants that were beginning to flower exhibited symptoms within 6 days. On the other hand, rarely did 15-day-old wild-type seedlings show symptoms. These results demonstrate that the lesion in jasmonate signaling affects susceptibility in the mutant throughout development and is not due simply to a difference in the timing of development between jar1-1 and wild-type plants.

Figure 4.

Effect of plant age on susceptibility to P. irregulare.

Numbers adjacent to plots indicate age of plants (days) at inoculation. For clarity, data for 31- and 39-days-old wild-type plants are not included as they showed no symptoms during the experiment. Values are the mean of two separate experiments with standard deviations (N = 14 or 15 pots/age).

To further test the role of jasmonate in resistance to Pythium irregulare we inoculated a fatty acid desaturase triple mutant essentially lacking the ability to synthesize jasmonate ( McConn et al. 1997 ). The results shown in Fig. 2 demonstrate that this mutant showed the same symptoms of susceptibility as jar1-1. Susceptibility of this jasmonate biosynthesis mutant further demonstrates the importance of jasmonate in Arabidopsis resistance to Pythium.

The jar1 locus maps near the distal end of chromosome 2

The jar1 locus was positioned on chromosome 2 based on its co-segregation with the phenotypic marker er in the Landsberg background (not shown). Mapping was refined by crossing jar1-1 to NW151, a line which has multiple phenotypic markers for chromosome 2, and isolating F2 individuals homozygous for jar1. Of 601 F3jar1 families scored for as1, 121 were heterozygous and three homozygous. Of 591 families scored for cer8, 19 were heterozygous and none were homozygous. Only four of the cer8–jar1 recombinants were also recombinant for as1, indicating the gene order was as1-jar1-cer8. The Kosambi mapping function ( Koorneef & Stam 1992) was used to calculate genetic map distances from jar1-1 of 11.4 c m (± 1.0) and 1.6 c m (± 0.4) for as1 and cer8, respectively.

Discussion

The evidence provided here further supports the fact that the JAR1 locus is specifically involved in jasmonate signaling, since ethylene, auxin and cytokinin response was not affected. Although sensitivity to inhibition of germination by ABA was previously found to be altered in jar1-1, the response was an increase rather than a decrease in sensitivity to ABA ( Staswick et al. 1992 ). While the relationship between ABA and jasmonate signaling pathways is not clear, a number of responses and genes are similarly affected by both ABA and MeJA. It is conceivable that interaction or ‘cross-talk’ occurs between these pathways, and thus a defect in one might affect the other.

Our new results also extend our previous finding that in addition to affecting root growth, the JAR1 locus is involved in jasmonate signaling responses in other parts of the plant ( Staswick et al. 1992 ). Suppression of jasmonate-mediated accumulation of VSP homologue mRNA in leaves agrees with our previous observation at the protein level and is consistent with results at the mRNA level reported by others ( Berger et al. 1996 ). The fact that the lesion in jar1-1 diminishes but does not eliminate gene expression could indicate that the mutation is leaky or that additional signaling pathways may be involved in expression of these genes.

Both linolenate and arachidonate induced VSP-homologue mRNA accumulation and, interestingly, jar1-1 plants were impaired in this response. This suggests that the induction of VSP by arachidonate in Arabidopsis leaves is mediated through a jasmonate signaling pathway, although a pleotropic effect cannot be ruled out. Linolenate may provide increased substrate for jasmonate biosynthesis, but it is unlikely that arachidonate (or linolenate) interacts directly with a jasmonate receptor. This is supported by the fact that both linolenate and arachidonate caused moderate reductions in root growth but, in contrast to jasmonate, there was no difference in response between mutant and wild type. Jasmonate stimulates a variety of stress responses in plants and may be a general signaling intermediate in these pathways ( Creelman & Mullet 1995;Sembdner & Parthier 1993;Staswick 1992). Our results suggest that arachidonate and linolenate activate a jasmonate-mediated stress response pathway in Arabidopsis leaves. Although VSP genes are induced by stress conditions in both soybean ( Staswick 1994) and Arabidopsis ( Berger et al. 1995 ), it is not known what role, if any, these proteins play in ameliorating adverse environmental conditions.

We determined that P. irregulare was the causal agent of the blighting in jar1-1 plants. As with most other species in the genus Pythium, P. irregulare is a common soil-inhabitant that often infects seedling roots, hypocotyls, and other juvenile tissues. It can also cause disease in mature plants depending upon host and environmental conditions ( Hendrix & Campbell 1973). It has been isolated from a wide range of crop and ornamental plant hosts ( van der Plaats-Niterink 1981) and is an economically important pathogen on such crops as carrot ( Liddell et al. 1989 ) and alfalfa ( Hancock 1991). It has also been reported on Brassica, a close relative of Arabidopsis (van der Plaats-Niterink 1981), but, to our knowledge, this is the first isolation of P. irregulare from Arabidopsis. We are currently investigating whether other isolates of P. irregulare and other species of Pythium are also infectious on our jasmonate response mutants.

Although we could infect mutant plants by distributing contaminated soil on the surface of pots, a more reproducible method of inoculation was to introduce small agar plugs containing the fungus into the root zone near the base of plants. We chose to assay plants as clusters of 15–20 seedlings because when inoculated individually some plants escaped infection (not shown). In contrast, inoculation of clusters usually resulted in infection of all inoculated units and nearly all plants in a unit showed symptoms. Although most mutant plants died after infection, some that initially showed symptoms survived and resumed growth, although their development was slowed considerably. This could indicate that the defect in jasmonate signaling is only a part of the overall resistance strategy. Alternatively, other environmental factors could also vary and influence susceptibility. Differential susceptibility was apparently not due to small differences in developmental stage since infection was not limited to young seedlings but also occurred in mature plants that were bolting, although the appearance of symptoms was somewhat delayed in older plants.

We have demonstrated that it is the jasmonate signaling lesion in jar1-1 and not a secondary mutation in another gene that results in susceptibility, since two additional jar1 alleles also confer susceptibility. The mutants also represent two different Arabidopsis ecotypes. The essential role of jasmonate in signaling resistance is further supported by the fact that a jasmonate biosynthesis mutant was also susceptible to P. irregulare. The fad triple mutant is also more susceptible to infestation by fungal gnats ( McConn et al. 1997 ). We have isolated an allele of the male sterile coi1 mutant and it was also susceptible to P. irregulare (data not shown). Although genes associated with response to micro-organism attack have been shown to be induced by jasmonate, our results provide direct evidence that jasmonate signaling in Arabidopsis is involved in resistance to a fungal pathogen.

In contrast to our results with Pythium, the coi1 mutant is more resistant to a coronatine-producing strain of the bacterial pathogen Pseudomonas syringae (Feys et al. 1994). Coronatine, having structural similarity to jasmonate, apparently predisposes Arabidopsis to infection by P. syringae. Thus coronatine/jasmonate insensitive coi1 mutants are less severely infected. The opposing relationship between jasmonate in disease resistance and coronatine in disease development is not yet understood.

In our hands, attempts to complement the lesion in the fad triple mutant by supplying MeJA exogenously prior to inoculation with P. irregulare, either as a spray to leaves or by drenching the soil, were unsuccessful. This approach restored resistance to fungal gnats in the fad mutant ( McConn et al. 1997 ). The quantity and timing of jasmonate application to roots may be critical in providing resistance to P. irregulare and meeting these conditions is complicated by the growth of roots in soil. We are attempting to address these potential problems by plant growth in other culture systems.

Asymptomatic colonization of wild-type roots by P. irregulare can be assumed ( Hendrix & Campbell 1973). Our results confirm that wild-type Arabidopsis plants are not immune per se to P. irregulare, as small numbers of wild-type plants did exhibit symptoms following inoculation. Although the role of jasmonate in resistance to P. irregulare is not known, several possible mechanisms come to mind based on our observations here and on what is known about jasmonate activity in plants. The fact that symptom development in wild-type plants occurred only in the youngest seedlings, and that symptoms appeared more rapidly in younger mutant plants suggests that pathways affected by the jar1 mutation were related to the conversion of tissues from a juvenile to a mature state, such as a reduction in root exudation or increase in lignification. Jasmonate is known to stimulate expression of genes of the phenylpropanoid pathway which are important for lignin biosynthesis ( Gundlach et al. 1992 ). On the other hand, the first symptom observed in affected plants was wilting, presumably due to the growth of the fungus into the vascular system or to girdling of the crown. Jasmonate has been implicated in drought tolerance ( Mason & Mullet 1990;Sembdner & Parthier 1993), and perhaps a decreased ability to compensate for loss of vascular function may be a factor in increased susceptibility in jar1. P. irregulare also causes a non-specific and localized necrosis or wounding of affected tissue. It is possible that resistance to P. irregulare is associated with wound response mechanisms already reported for jasmonate, such as flavonoid ( Gundlach et al. 1992 ) or steroid glycoalkaloid ( Choi et al. 1994 ) biosynthesis. Jasmonate is also known to stimulate the production of antifungal proteins such as thionins ( Andresen et al. 1992 ) and osmotin ( Xu et al. 1994 ) in plants. Finally, proteinaceous toxins and cellulytic enzymes have been implicated in tissue necrosis by P. irregulare (reviewed in Martin 1994). It is possible that the signaling pathway defined by jar1 leads to the production of proteins or other molecules that inhibit or inactivate specific toxins or cellulytic enzymes.

The jar1 locus maps to the lower end of chromosome 2 between phenotypic markers as1 and cer8. Interestingly, the coi1 locus also maps between these markers. However, in contrast to jar1 it is much closer to as1 than to cer8 (Xie et al. 1998). Therefore, it is clear that JAR1 and COI1 are distinct genes. This mapping information has allowed us to make use of the available Arabidopsis molecular marker maps to refine the location of jar1 and to screen BAC libraries for candidate clones containing the gene. Identification and characterization of the JAR1 gene will lead to a better understanding of the mechanism of jasmonate signaling in plants. During revision of this manuscript Vijayan et al. (1998) reported increased susceptibility of the fad triple mutant to P. mastophorum, indicating that a role for jasmonate in resistance is not specific for P. irregulare. Further study of the interaction between jasmonate signaling mutants and Pythium species, as well as other micro-organisms, will further our understanding of the mechanism of disease resistance in general, and more specifically in this economically important genus of pathogens.

Experimental procedures

Plant material

Plants for RNA isolation were grown in continuous light in Magenta boxes in 50 ml of liquid Murashige-Skoog (MS) basal salt medium ( Murashige & Skoog 1962) obtained from Sigma. Tissue was harvested by blotting away excess liquid, separating leaves from roots and then freezing at – 80°C until RNA isolation. The jar1-1 allele was previously described and found to be a recessive trait ( Staswick et al. 1992 ). Two new alleles of jar1 were isolated by screening M2 populations as described earlier ( Staswick et al. 1992 ). The jar1-2 allele was isolated from a population of ecotype Columbia irradiated with 50 kRad from a 60Co source. jar1-4 was found in a T-DNA mutagenized pool (CS3115) (ecotype Wassilewskija) obtained from the Arabidopsis Biological Resource Center (Ohio State University). The fatty acid desaturase (fad) triple mutant was kindly provided by J. Browse.

Assay for root growth

Seedling root growth was assayed on MS medium agar plates as described previously ( Staswick et al. 1992 ), except that the agar concentration was 1% w/v. Filter sterilized MeJA or other compounds were added to the medium after autoclaving at the concentrations indicated for each experiment. MeJA was obtained from Bedoukian Research, Danbury, CT, USA. Jasmonic acid was obtained from the methyl ester by saponification as outlined by Dathe et al. (1981) .

RNA isolation and assay

Total RNA was isolated by an acid guanidinium thiocyanate-phenol-chloroform extraction protocol ( Chomczynski & Sacchi 1987) using ‘TRIzol’ reagent according to the manufacturer’s procedures (GibcoBRL, Bethesda, MD, USA). RNA was fractionated by electrophoresis, blotted and hybridized using materials from a ‘Northern Max’ blotting kit as specified by the manufacturer (Ambion Inc., Austin, TX, USA). Hybridization was at 42°C and washes were in 0.2× SSC at 62°C. Figures of hybridizations were prepared by acquiring images from X-ray films with a Microtek ScanMaker III flatbed scanner and then assembling images with Adobe Photoshop.

Probes for hybridization were generated with 32P dCTP by random primer labeling using gel purified cDNA inserts as described earlier ( Staswick 1997). The Arabidopsis VSP homologue was from EST clone 108B11T7, obtained from the Arabidopsis Biological Resource Center.

Isolation and identification of Pythium irregulare

Entire mutant plants exhibiting disease symptoms were extricated from pots and washed under running water for 15 min to remove planting medium and debris. The washed plants were blotted dry and placed on water agar in Petri dishes. Asymptomatic plants were also thus cultured. The cultures were incubated for up to 3 days at room temperature and examined daily for fungal growth on plant surfaces and in the agar. To culture fungi from planting medium, 25 g of medium was suspended in 100 ml sterile distilled water and 1 ml volumes of suspension was applied to the surface of water agar plates.

Fungi, identified as Pythium by colony and hyphal morphology, grew in near-pure culture from all diseased plant material by 2 days and was detected in cultures of the planting medium. The fungus was absent from healthy plants. Hyphal tip transfers were made of fungi growing from the diseased plants onto potato dextrose agar. Three isolates were purified and transferred to dishes containing boiled grass blades in sterile water. After 10 days of growth, the fungus in each grass blade culture was examined microscopically for morphology of the vegetative and reproductive structures. The three isolates were identical. Oogonia (mean diameter 16 μm, range 14–18 μm) were terminal, globose, and with one or two short, blunt projections. Oospores were aplerotic, with walls approximately 1.5 μm thick. Antheridia were diclinous. Hyphal swellings (mean diameter 16 μm, range 14–18 μm) were intercalary and limoniform to globose. No zoospores were produced. These characteristics are consistent with P. irregulare as described in van der Plaats-Niterink (1981). There were no obvious differences in morphology and growth rate between the isolates from Arabidopsis and P. irregulare isolates from other hosts in G. Yuen’s collection.

Growth of P. irregulare and inoculation of plants

Cultures of P. irregulare were grown on potato dextrose agar with transfers to fresh media made every 4–6 weeks. Plants were grown in Redi-Earth (Scotts, Marysville, OH, USA) in 7 cm square pots in a growth chamber at 21°C and 9 h day length. Each pot contained clusters of 10–20 seedlings, one of wild-type and one or more mutant lines, depending on the experiment. At the indicated times each plant cluster was inoculated with 5 agar plugs (1 mm diameter) taken from a 6–7-day-old P. irregulare culture using a Pasteur pipette. The agar plugs were placed 2–4 cm below the soil surface and around the perimeter of each seedling cluster. Controls were mock inoculated in the same way except that the agar plugs were taken from plates lacking the fungus. After inoculation, plants were returned to the growth chamber and monitored daily for symptoms of loss of turgor and tissue collapse.

Mapping the jar1 locus

The jar1-1 mutant was crossed to Landsberg er and the F2 generation was screened for the recessive long root phenotype on agar plates containing 50 μm MeJA. Mutants that were rescued to soil and grown to maturity co-segregated with er, indicating the jar1-1 locus was on chromosome 2. jar1-1 was then crossed to the chromosome 2 marker line NW151 (Notingham Arabidopsis Stock Center) and the F2 generation was screened for jasmonate resistance as outlined above and seed was collected from these. F3 families of at least 50 seeds were planted and monitored for the expected phenotypic marker traits. The as1 phenotype was evident in young plants which had asymmetric leaves and the cer8 phenotype appeared as dark glossy green inflorescence stalks, rather than the dull waxy appearance of wild type.

Acknowledgements

This work was supported in part by NSF-EPSCoR Cooperative Agreement EPS-9255225. This is Journal Series Paper no. 12172 from the Nebraska Agriculture Research Division. We thank Dr J. Browse for providing the fatty acid desaturase mutant and Dr M. Dickman for helpful advice.

Ancillary