Abstract
We identified an Arabidopsis (Arabidopsis thaliana) mutant, sma4 (symptoms to multiple avr genotypes4), that displays severe disease symptoms when inoculated with avirulent strains of Pseudomonas syringae pv tomato, although bacterial growth is only moderately enhanced compared to wild-type plants. The sma4 mutant showed a normal susceptible phenotype to the biotrophic fungal pathogen Erysiphe cichoracearum. Significantly, the sma4 mutant was highly resistant to a necrotrophic fungal pathogen, Botrytis cinerea. Germination of B. cinerea spores on sma4 mutant leaves was inhibited, and penetration by those that did germinate was rare. The sma4 mutant also showed several pleiotropic phenotypes, including increased sensitivity to lower humidity and salt stress. Isolation of SMA4 by positional cloning revealed that it encodes LACS2, a member of the long-chain acyl-CoA synthetases. LACS2 has previously been shown to be involved in cutin biosynthesis. We therefore tested three additional cutin-defective mutants for resistance to B. cinerea: att1 (for aberrant induction of type three genes), bodyguard, and lacerata. All three displayed an enhanced resistance to B. cinerea. Our results indicate that plant cutin or cuticle structure may play a crucial role in tolerance to biotic and abiotic stress and in the pathogenesis of B. cinerea.
Disease resistance in plants is mediated by both preformed and induced defenses. Examples of preformed defenses include physical barriers such as the cutin layer on leaf and stem surfaces; antimicrobial enzymes stored in the tonoplast that are released upon cell damage; and secondary metabolites that are often present at low constitutive levels, but whose levels are increased during pathogen attack (Morrissey and Osbourn, 1999). Induced defenses include production of reactive oxygen species, secretion of antimicrobial proteins, cross-linking of cell walls, production of additional secondary metabolites, and the hypersensitive response (HR; a type of programmed cell death; Glazebrook, 2005). Rapid induction of defenses is usually mediated by disease resistance (R) genes, which encode proteins that detect specific pathogen virulence proteins (Jones and Dangl, 2006). The hallmark of R gene-mediated resistance is specificity. Most R proteins detect only one or two virulence proteins. However, different R genes activate very similar responses, suggesting that there are convergent points in R protein signaling. Despite this seeming convergence, we have a very poor understanding of what proteins function downstream of R proteins to activate the HR and other defense responses.
A large number of R genes from plants and their cognate virulence genes from pathogens have been cloned. In Arabidopsis (Arabidopsis thaliana), the R genes RPM1 and RPS2 have been intensively studied (Bent et al., 1994; Grant et al., 1995; Warren et al., 1998). RPM1 mediates detection of at least two different pathogen virulence proteins, AvrB and AvrRpm1 (Bisgrove et al., 1994), while RPS2 confers recognition of AvrRpt2 (Kunkel et al., 1993). In an attempt to identify genes that function downstream of R gene activation, we designed a forward genetic screen to identify Arabidopsis mutants that became susceptible to a Pseudomonas syringae pv tomato (Pst) strain that carried both avrRpt2 and avrB [Pst DC3000(avrRpt2, avrB)]. The wild-type Arabidopsis variety Columbia (Col)-0 contains both RPS2 and RPM1. We hoped to avoid recovering rpm1 or rps2 mutants, since in theory mutations in either gene alone should still be resistant to such a strain.
We expected to identify several classes of mutants from this screen. The first class of mutants was expected to contain mutations in downstream signaling components that are shared by both RPM1 and RPS2, such as positive regulators of defense responses, including transcription factors and regulators of the HR. A second class of mutants was expected to be in genes required for assembly and/or stability of R protein complexes such as SGT1 and RAR1 (Tornero et al., 2002). A third class of mutants was expected to contain mutations in genes that generally make the plant more susceptible to both virulent and avirulent pathogens, such as mutations in preformed defense systems or components of the salicylic acid signaling pathway.
As described below, we uncovered mutations in the first and third class. In this article, we describe characterization and cloning of enhanced susceptibility mutant sma4 (symptoms to multiple avr genotypes4), which appears to belong to the third class. The sma4 mutant displayed severe disease symptoms after infection with Pst DC3000 carrying avrRpt2 and avrB, either together or individually. Interestingly, this mutant was highly resistant to the necrotrophic fungal pathogen Botrytis cinerea. We cloned the SMA4 gene by map-based cloning, and found that SMA4 encodes a member of the long-chain acyl-CoA synthetase (LACS) family, LACS2, which has previously been shown to function in cutin synthesis (Schnurr et al., 2004).
RESULTS
Identification of Arabidopsis Mutants Susceptible to a P. syringae Strain Expressing Both avrB and avrRpt2
Approximately 16,600 ethylmethanesulfonate, diepoxybutane, and fast neutron-mutagenized M2 plants were screened for disease symptoms following inoculation with DC3000 harboring both avrB and avrRpt2 on separate plasmids. Nine putative mutants that displayed strong symptoms were selected. M3 progeny of these nine mutants were dip inoculated with strain DC3000 expressing avrB or avrRpt2 individually. Three of the mutants retained resistance to DC3000(avrB) but appeared fully susceptible to DC3000(avrRpt2). Allelism tests revealed that these mutations were in fact in RPS2. Although this was an unexpected result at the time, we now know that AvrRpt2 affects RPM1 function by eliminating RIN4 (Axtell and Staskawicz, 2003; Mackey et al., 2003), a plant protein required for RPM1-mediated resistance (Mackey et al., 2002). Thus, rps2 mutants become susceptible to DC3000(avrRpt2, avrB) because AvrRpt2 eliminates RIN4, blocking RPM1 function.
M3 progeny of the remaining six mutants displayed disease symptoms when inoculated with either DC3000(avrB) or DC3000(avrRpt2). Allelism tests of these six mutants indicated that three of the mutations were allelic to ndr1 (Century et al., 1997). NDR1 is a membrane-associated protein required by several CC-NBS-LRR class R genes, including RPM1, RPS2, and RPS5 (Century et al., 1995).
The remaining three mutations were not allelic to ndr1 or each other. We designated these mutants sma1, sma3, and sma4. Here, we describe characterization of the sma4 mutant and the cloning of the SMA4 gene. The other two mutants remain to be fully characterized.
The sma4 Mutation Confers Partial Susceptibility to Several Avirulent P. syringae Strains
The sma4 mutant displays enhanced susceptibility to Pst DC3000 expressing avrRpt2 or avrB. A and B, Six-week-old sma4 plants were dip inoculated with Pst DC3000(avrRpt2) (A) or Pst DC3000(avrB) (B) and photographed 3 d later. C and D, Bacterial growth for Pst DC3000(avrRpt2) (C) or Pst DC3000(avrB) (D) was monitored over a 4-d time course. Each data point represents the mean and se of three samples.
The symptoms of necrotic leaf collapse in response to each of the Pst DC3000 avr genotypes were very atypical compared to those observed when virulent DC3000 was allowed to infect wild-type Col-0 plants. The sma4 symptoms in response to infection by strain DC3000(avrRpt2) were the most severe. Typically, some necrotic collapse was observed as early as 48 h after dip inoculation at 1.5 × 108 cfu/mL. By late day 3 and early day 4 after infection, most of the older, outer rosette leaves were observed to be entirely collapsed and dead. In contrast, a virulent Pst DC3000 infection on wild-type Col-0 plants typically produces chlorosis, large water-soaked lesions, and pin-point necrotic pits over the surface of Arabidopsis leaves (Whalen et al., 1991). Rarely is whole leaf collapse and death observed in a virulent Pst DC3000 interaction at an inoculum concentration of 1.5 × 108 cfu/mL. Strains of DC3000 carrying avrB, avrRpm1, and avrPphB induced a phenotype on sma4 plants similar to that induced by DC3000(avrRpt2). However, these responses were consistently less severe (Fig. 1; data not shown).
Examination of sma4 plants inoculated with the same concentration of the virulent control strain DC3000(avrB∷Ω) (Simonich and Innes, 1995) consistently showed symptoms of chlorosis, water-soaked lesions, and necrotic pits, similar to wild-type Col-0. This indicated that sma4 plants could develop classic symptoms of susceptibility. However, the disease symptoms appeared more damaging to the sma4 plants than to Col-0 plants. Often, the sma4 mutant plants died within 4 to 5 d of infection by strain DC3000(avrB∷Ω), while identical inoculations of wild-type Col-0 plants were never observed to kill the plants (data not shown). Interestingly, although sma4 displayed more severe disease symptoms, there was no significant difference in bacterial growth between Col-0 and sma4 plants 3 d after infection by strain DC3000(avrB∷Ω) (data not shown).
The sma4 phenotype of necrotic collapse more closely resembled a HR than true susceptibility. The HR is strongly associated with cessation of bacterial growth and other resistant responses in the infected plant (Klement, 1982). To determine whether the atypical symptoms observed on sma4 plants reflected susceptibility or an extreme hypersensitive resistance, growth of two different DC3000 avirulent genotypes was monitored in sma4 leaves. Figure 1, C and D, shows that DC3000(avrRpt2) and DC3000(avrB) grew to higher levels in sma4 leaves compared to wild-type Col-0 leaves, but this growth was not as high as that of the virulent control strain DC3000(avrB∷Ω). These results indicate that the sma4 mutant phenotype is not the result of an enhanced HR. They also indicate that the sma4 mutation does not block signaling by the RPM1 and RPS2 R genes, as growth of the avirulent strains is still significantly reduced.
sma4 Leaves Are Intolerant of Weak Salt Solutions
We normally use a low (1 × 105 cfu/mL) concentration of pathogen carried in a solution of 10 mm MgCl2 with 0.001% L-77 wetting agent as the method of inoculation for monitoring in planta pathogen growth. However, we noted that sma4 plants rapidly (within 2 h) exhibited severe necrotic leaf collapse after vacuum infiltrating pathogen cells suspended in a 10 mm MgCl2 solution. Vacuum infiltration causes the leaves to become saturated with the infiltrated solution in the intercellular leaf spaces. The plants are allowed to shed the excess water by drying out slowly over a period of a few hours. This drying is occasionally observed to damage some wild-type leaves along their margins, where some necrosis may develop. However, sma4 plants consistently developed extensive tissue collapse on more than 50% of their leaves or were entirely killed by the drying process. The collapse developed too quickly to be explained by a pathogen-induced effect such as HR induction, which normally takes at least 12 h to become visible. We assessed whether the vacuum infiltration process, the subsequent leaf drying, or the buffer components were the cause of this sma4 phenotype. Vacuum infiltration of sterile 10 mm MgC12 and 0.001% L-77 caused the same degree of sma4 tissue collapse as was seen with the pathogen inoculations. Vacuum infiltration with the pathogen carried in distilled water plus 0.001% L-77 did not cause any more sma4 leaf collapse than collapse of wild-type Col-0 leaves within 2 h of the infiltration. Finally, sma4 leaves vacuum infiltrated with distilled water plus L-77 and no pathogen also did not display collapse. These observations suggested that the 10 mm MgCl2 was responsible for the leaf collapse. We therefore used distilled water instead of 10 mm MgCl2 to generate the bacterial growth curve data shown in Figure 1.
To ascertain whether this effect was limited to salt exposure or reflected a general osmotic sensitivity conferred by the sma4 mutation, we repeated the vacuum infiltrations with sterile 15 mm NaC1, 15 mm MgSO4, sterile 20 mm Glc, and sterile 20 mm mannitol plus 0.001% L-77 wetting agent. The NaCl and MgSO4 treatments produced a response similar to the MgCl2 treatments in severity, while the Glc and mannitol treatments responded similar to infiltration with water. We also noted that we could block the observed collapse by loosely sealing the salt-infiltrated sma4 plants under plastic covers, which kept the surrounding humidity relatively high. However, these covers did not allow the leaves to dry by shedding the excess water from their intercellular spaces. These results suggested that the rapid tissue collapse (within 2 h) was triggered by salt exposure, but also required the drying process after salt exposure. The phenotype was not attributable to a general osmotic sensitivity of the sma4 leaf cells.
sma4 Seedlings Are Highly Sensitive to Lowered Humidity
The sma4 mutant is highly sensitive to lowered humidity. Wild-type Col-0 and sma4 plants were grown in soil for 10 d after germination under a clear plastic dome to maintain near 100% relative humidity. A, Seedlings photographed immediately after removing the dome. B, Seedlings photographed 2 h after the dome was removed.
sma4 Leaf Cells Leak Ions Faster Than Wild-Type Col-0 Cells during the HR and after Salt Exposure
Ion leakage from sma4 mutant leaves is enhanced. A, Col-0 and sma4 ion leakage in response to PsgR4(avrRpt2). PsgR4(avrRpt2) was infiltrated into single leaves at a concentration of 5 × 107 cfu/mL. Five leaf discs were floated abaxial side down in deionized water for 4 h, after which conductivity of the solution was measured in micro-ohms. Each data point is the mean ± se of three samples. B, sma4 leaves leak ions faster than wild-type Col-0 leaves after salt exposure. Ion leakage from sma4 and Col-0 leaves was measured after vacuum infiltration with MgCl2 or MgSO4 solution.
We also assessed the effect of vacuum infiltration of 10 mm MgC12 and 10 mm MgSO4 on ion leakage (Fig. 3B). The rapid collapse of sma4 leaves correlated with a greater degree of ion leakage from sma4 leaves after salt treatment compared to Col-0 leaves (Fig. 3B).
sma4 Plants Display Enhanced Disease Resistance to B. cinerea
The sma4 mutant displays enhanced disease resistance to B. cinerea. A, Leaves from 6-week-old Col-0 and sma4 plants were detached and placed in petri dishes and inoculated with B. cinerea. Leaves were photographed 5 d after inoculation. B, Lesion size induced by B. cinerea. Lesion size was determined by measuring the major axis of the necrotic area at 5 d after inoculation. Bars represent the mean and sd from eight samples. C and D, Leaves from Col-0 and sma4 plants were stained with trypan blue and photographed 3 d after infection with B. cinerea.
In addition to B. cinerea, we also tested a biotrophic fungal pathogen, Erysiphe cichoracearum strain UCSC, which is virulent on wild-type Col-0 plants (Adam and Somerville, 1996), and another necrotrophic fungal pathogen, Alternaria brassicicola. We observed no significant difference between Col-0 wild type and the sma4 mutants in their response to these pathogens (data not shown).
sma4-Mediated Resistance to B. cinerea Is COI1 and EIN2 Independent
Resistance to B. cinerea does not require COI1 or EIN2. A, Leaves from 6-week-old plants were detached and placed in petri dishes and inoculated with B. cinerea. Representative leaves were photographed 3 d after inoculation. B, Lesion size induced by B. cinerea was measured 3 d after inoculation. Bars represent the mean and sd from eight samples. C, Six-week-old plants were dip inoculated with Pst DC3000(avrRpt2). Plants were photographed 3 d after inoculation.
Genetic Mapping and Identification of SMA4
To characterize the genetic basis of sma4-mediated responses to pathogens, the mutant was backcrossed to wild-type Col-0. The Fl progeny were resistant to dip inoculation with DC3000(avrRpm1), which indicated that sma4 was a recessive mutation. Consistent with this, the F2 progeny segregated 3:1 wild-type:sma4 mutant phenotypes. To verify that partial susceptibility to multiple avr genotypes was indeed conferred by a single mutation, several F3 families from the sma4 × Col-0 F2 progeny were tested for symptoms after dip inoculation with DC3000(avrRpm1), DC3000(avrB), and DC3000(avrRpt2). These F3 families displayed symptoms of partial susceptibility to each of the different avr genotypes, indicating that the sma4 phenotype was conferred by a single mutant locus.
Structure of the SMA4 gene and complementation of the sma4 mutation. A, Exon and intron structure of SMA4. The exon regions are indicated with rectangles and intron regions with lines. The sma4 mutation is a point mutation in intron 4. The letters show nucleotide sequences from the beginning of intron 4. The T-to-A mutation in sma4 is indicated by an uppercase letter. B and C, Complementation of the sma4 mutation by transformation. Plants were dip inoculated with DC3000(avrRpt2) and photographed 3 d after inoculation (B), or inoculated with B. cinerea and photographed 5 d after inoculation (C).
To further confirm that SMA4 is At1g49430, we complemented the sma4 mutant with a 6.8-kb genomic DNA sequence containing the full-length At1g49430 gene including 1.5 kb of upstream promoter region and 0.8 kb of downstream sequences. Figure 6, B and C, shows that this genomic DNA fragment rescued sma4-mediated disease phenotypes. These data demonstrated that At1g49430 is the SMA4 gene.
SMA4 Encodes an Acyl-CoA Synthetase
At1g49430 has previously been shown to encode a LACS, of which there are nine family members in Arabidopsis (Shockey et al., 2002). Specifically, At1g49430 encodes LACS2, which has been shown to be involved in cutin biosynthesis (Schnurr et al., 2004). We tested lacs2-1, a T-DNA insertion allele (Schnurr et al., 2004), for its response to Pst DC3000(avrRpt2) and found that it displayed a sma4-like susceptible phenotype (Fig. 6B). We also tested the lacs2-1 mutant for resistance to B. cinerea and found it to display sma4-like enhanced resistance (Fig. 5B), confirming that these phenotypes in the sma4 mutant are likely caused by the loss of LACS2 function.
It should be noted that the lacs2-1 mutant has been reported to display a strong dwarfed phenotype with small wrinkled leaves (Schnurr et al., 2004), which we did not observe in the sma4 mutant. Under our growth conditions, however, lacs2-1 plants were only slightly smaller than sma4 and wild-type plants (Fig. 6B; data not shown), suggesting that the dwarfed phenotype reported by Schnurr et al. (2004) is influenced by environmental conditions such as daylength, light intensity, and relative humidity. The slight decrease in plant size that we did observe in the lacs2-1 mutant might be caused by differences in genetic background, as the lacs2-1 mutant was isolated in the Col-0 glabrous-1 mutant background while sma4 was isolated in the Col-0 wild-type background.
Several Different Arabidopsis Mutants with Altered Cuticle Structure Display Enhanced Resistance to B. cinerea
The cutin-defective mutants att1, bdg, and lcr display enhanced resistance to B. cinerea. A, Leaves from 5-week-old plants were detached and placed in petri dishes and inoculated with B. cinerea. Leaves were photographed 3 d after inoculation. B, Lesion size induced by B. cinerea. Lesion size was determined by measuring the major axis of the necrotic area at 3 d after inoculation. Bars represent the mean and sd from eight samples.
DISCUSSION
We have shown that loss of LACS2 function makes Arabidopsis more susceptible to avirulent strains of the Arabidopsis bacterial pathogen Pst DC3000, but more resistant to a virulent strain of the necrotrophic fungus B. cinerea. Why does loss of LACS2 function cause these phenotypes? LACS2 encodes a LACS (Schnurr et al., 2004). LACS enzymes activate free fatty acids to acyl-CoAs, a key step for the utilization of fatty acids by most lipid metabolic enzymes (Shockey et al., 2002). LACS2 is expressed in the epidermal cell layer of Arabidopsis leaves and the T-DNA insertion mutant lacs2-1 has a thinner cuticle layer compared to wild-type plants. A preferred substrate of the LACS2 enzyme is ω-hydroxypalmitic acid; thus, Schnurr and colleagues concluded that LACS2 catalyzes the synthesis of ω-hydroxy fatty acid-CoA intermediates required for cutin synthesis (Schnurr et al., 2004). Leaves of a lacs2-1 null mutant release chlorophyll faster than wild-type leaves when immersed in 80% ethanol and support pollen germination, which indicates that the cuticular barrier on lacs2-1 plants is more permeable (Schnurr et al., 2004). We propose that it is this increased permeability that causes the enhanced tissue collapse upon infiltration of dilute salt solutions or avirulent pathogens, and that causes enhanced resistance to B. cinerea.
Support for this hypothesis comes from work on ATT1, which encodes an Arabidopsis Cyt P450 monooxygenase (CYP86A2) that is required for proper cuticle development. Loss-of-function att1 mutants have lower cutin content, loose cuticle ultrastructure, and increased rates of water vapor transmission (Xiao et al., 2004). Because cutin also lines the substomatal chamber in Arabidopsis leaves, this change in ultrastructure likely allows more rapid leakage of cell fluids into this area, which is a primary point of colonization during P. syringae infection. An increase in water flow into the substomatal chamber would be expected to increase the growth of P. syringae strains as water stress is believed to be a limiting factor during P. syringae infections (Wright and Beattie, 2004). Consistent with this hypothesis and similar to the sma4 mutant, the att1 mutant displays enhanced disease symptoms, including leaf collapse, when inoculated with Pst DC3000. Furthermore, Xiao et al. showed that the expression of the P. syringae virulence genes avrPto and hrpL is greatly enhanced in the substomatal chambers of the att1 mutant compared to wild-type plants. This observation suggests that the att1 mutant produces an eliciting signal that accumulates in the substomatal chamber more readily than in wild-type plants or, alternatively, fails to produce a signal that suppresses induction of these genes. The former hypothesis is consistent with a reduction in the cuticular barrier lining the substomatal cavity.
An increase in cuticle permeability to water vapor in the sma4 mutant is consistent with our observation that sma4 seedlings are quite sensitive to rapid decreases in humidity (Fig. 2). An increase in cuticle permeability would also explain the observed increase in ion leakage after bacterial infection or salt exposure (Fig. 3).
The most interesting phenotype of sma4, however, is its resistance to the necrotrophic fungus B. cinerea. Given that sma4 leaves collapse and die more rapidly when infected with P. syringae, an increased resistance to a necrotroph was unexpected. One possible explanation for the enhanced resistance is that B. cinerea may rely on specific physical and/or chemical queues on the leaf surface to promote germination and penetration. Host surface structure has been implicated in the pathogenesis of other necrotrophic pathogens. For example, in the interaction between Colletotrichum trifolii and alfalfa (Medicago sativa), the host surface chemistry appears to be important for induction of the fungal gene expression required for pathogenic development (Dickman et al., 2003). Under this scenario, because of the inappropriate host cuticle structure in the sma4 mutant, B. cinerea is unable to induce gene expression required for pathogenic development. As a result, the spores fail to penetrate the host surface, and, ultimately, fail to colonize and cause disease in the host plants. Arguing against this hypothesis, however, is that B. cinerea has a very broad host range, infecting over 200 plant species, which likely have different surface chemistries. In addition, B. cinerea is routinely cultivated on potato dextrose agar, on which spores readily germinate, indicating that B. cinerea does not have special requirements for germination, whereas germination on sma4/lacs2 mutant leaves was poor. This latter observation suggests that something on sma4/lacs2 mutant leaves may actively inhibit germination and hyphal growth.
If the inhibition hypothesis is correct, it is plausible that the permeability of the cuticle enhances export of an antifungal compound to the leaf surface or, alternatively, enhances import of a fungal elicitor that triggers production of an antifungal compound. Evidence that export of antifungal compounds to the leaf surface is important comes from recent work on the Arabidopsis PEN3 gene. PEN3 encodes a plasma membrane-localized ATP-binding cassette transporter that localizes around fungal penetration sites (Stein et al., 2006). Loss of PEN3 function causes increased susceptibility to the necrotrophic fungus Plectosphaerella cucumerina (Stein et al., 2006), implying that export of an unidentified compound contributes to resistance to this fungus. However, pen3 mutants do not display enhanced susceptibility to B. cinerea, possibly because wild-type Arabidopsis is already quite susceptible. Further evidence that increases in cuticle permeability lead to enhanced resistance to B. cinerea comes from our finding that three different cuticle-defective mutants all display enhanced resistance to B. cinerea (Fig. 7). These mutants have different structural changes in their cutin, but all have increased permeability (Wellesen et al., 2001; Xiao et al., 2004; Kurdyukov et al., 2006).
It has recently been reported that a transgenic Arabidopsis line that expresses a fungal cutinase displays enhanced resistance to B. cinerea (Chassot and Métraux, 2005; Chassot et al., 2007). Similar to lacs2 mutants, the cutinase transgenic lines form a defective cuticle that has increased permeability (Sieber et al., 2000). Most significantly, these transgenic lines, as well as the bdg mutant, were shown to release a fungitoxic activity from the surface of their leaves that was not released by wild-type leaves (Chassot et al., 2007). Thus, the enhanced resistance to B. cinerea observed in sma4 plants is likely due to this same antifungal activity. This supposition has very recently been confirmed by Bessire et al. (2007), who have shown that diffusates from lacs2 mutant leaves contain a strong antifungal activity. These observations suggest that Arabidopsis produces a compound that is very toxic to B. cinerea, but that is limited in its effectiveness either by its ability to reach the leaf surface in a timely manner or to be induced in a timely manner. Increasing the permeability of the cuticle layer, by any means, enhances its induction and/or release, thus conferring resistance.
MATERIALS AND METHODS
Plant Growth Conditions and Mutant Screening
Arabidopsis (Arabidopsis thaliana) plants were grown in growth rooms under a 9-h-light/15-h-dark cycle at 23°C as described previously (Frye and Innes, 1998). Approximately 16,600 ethylmethanesulfonate, diepoxybutane, and fast neutron-mutagenized Col-0 plants (M2 generation) were inoculated with Pseudomonas syringae pv tomato (Pst) strain DC3000 carrying both avrB (on plasmid pDSK600) and avrRpt2 (on plasmid pVSP61) and scored for disease responses 3 d after inoculation. Plants displaying severe disease phenotypes were selected and allowed to set seeds. The sma4 mutant was backcrossed twice to wild-type Col-0 prior to performing the analyses presented in this article, with the exception of the original bacterial growth data shown in Figure 1, C and D, which were performed on nonbackcrossed material.
Infections with Pathogens
Inoculation of Arabidopsis plants with Pst DC3000 and measurement of bacterial growth within leaves was performed as described previously using vacuum infiltration of 6-week-old plants (Simonich and Innes, 1995), except that inocula were prepared in distilled water rather than 10 mm MgCl2 to avoid the confounding effects of sma4 tissue damage caused by salt solutions. HR inoculations were performed by injecting bacterial cells suspended in distilled water into the abaxial side of 6-week-old leaves with a needleless 1-mL syringe. HR formation was assayed 24 h after infiltration. Inoculation of Arabidopsis plants with Botrytis cinerea was performed as described by Ferrari et al. (2003) using detached leaves (strain obtained from F.M. Ausubel and described by Ferrari et al. [2003]). Fungal structures and dead plant cells were stained by trypan blue (Frye and Innes, 1998). Samples were observed and photographed using a Nikon SMZ1500 dissecting microscope. Erysiphe cichoracearum strain UCSC1 was maintained and inoculated onto Arabidopsis plants as described previously (Tang and Innes, 2002).
Ion Leakage Measurements
Intact leaves were entirely infiltrated with a bacterial density of 5 × 107 cfu/mL in deionized water. At 0, 5, 10, 15, 20, and 25 h after infiltration, a random sample of inoculated leaves was excised from the plants and five leaf discs, each 7 mm in diameter, were cut from the excised leaves with a cork borer and immediately floated, abaxial side down, in 6 mL of 0.001% L-77 surfactant (Union Carbide) in water. The sampling was done in triplicate for each time point. Leaf discs were floated for 4 h, after which conductivity measurements of the bathing solution were made with a Radiometer Copenhagen CDM2f conductivity meter and Radiometer Copenhagen CDC104 detector (The London Company).
Genetic and Physical Mapping of sma4
F2 progeny of a sma4 cross to Landsberg erecta were used to genetically map the SMA4 gene. The F2 plants were inoculated with Pst DC3000(avrRpt2) and scored 3 d after inoculation. Plants displaying a sma4 phenotype were used for mapping. Initially, the sma4 mutation was mapped to chromosome I between SSLP markers T27K12 and CIW1. To further localize the SMA4 gene, we developed new markers at intervals between these two markers using Monsanto Col-0 and Landsberg erecta polymorphism data (marker data available upon request). A total of 556 susceptible F2 plants representing 1,112 meioses were scored, which enabled us to localize sma4 between two markers at positions 29 kb and 117 kb of the bacterial artificial chromosome (BAC) clone F13F21 (GenBank accession no. AC007504), defining an 88-kb region that cosegregated with the sma4 mutation.
Sequencing of Candidate Genes
Candidate genes were amplified by PCR from genomic DNA isolated from the sma4 mutant and from wild-type plants and directly sequenced. All sequencing reactions were performed using BigDye Terminator kits (Applied Biosystems) and separated on an ABI 3730 automated DNA sequencer. In total, 15 genes were amplified from the sma4 mutant by PCR and directly sequenced. To obtain the SMA4 cDNA sequence, RNA was isolated and first-strand cDNA synthesis was performed as described previously (Tang et al., 2005).
Complementation of the sma4 Mutant
A full-length SMA4 genomic sequence, including the promoter region and 3′ untranslated region, was amplified from BAC F13F21 using the following primers: 5′-AACCGCTAGCTTCCTTATAAAAAGTTAAAGAAAAAG-3′ and 5′-AATTGGGCCCCGTATGAGAATGATTAGTTTAGTTGA-3′. The PCR product was digested with ApaI and NheI and ligated with the binary vector pGreen0029 digested with ApaI and XbaI (Hellens et al., 2000). This DNA sequence contains a full-length At1g49430 gene including 1.5 kb of sequence 5′ to the start codon and 0.8 kb of sequence 3′ from the stop codon.
The pGreen0029:At1g49430 construct was transformed into Agrobacterium tumefaciens strain GV3101 by electroporation and selected on LB plates containing 50 μg/mL kanamycin sulfate (Sigma). Arabidopsis plants were transformed using the floral dip method (Clough and Bent, 1998). Transgenic plants were selected by growing on 0.5× Murashige and Skoog salts plus 0.8% agar and 50 μg/mL kanamycin. Transformants were transplanted to soil 7 d after germination and were inoculated with Pst DC3000(avrRpt2) and with B. cinerea when 5 weeks old.
Construction of Double Mutants
Double mutants were created by standard genetic crosses. The ein2-1 mutation was in the Col-0 background (Alonso et al., 1999), and coi1-1 was in the Col-6 background (Xie et al., 1998). To identify the sma4 mutation in F2 progeny, we performed PCR amplification using dCaps primers 5′-TATGTTATGACATGATCCAATCAATC-3′ and 5′-AATTTTAGATATAGGTCAATTTTTTTGT-3′, followed by digestion with RsaI (New England Biolabs).
ACKNOWLEDGMENTS
We thank J. Browse for providing lacs2-1 seeds, J. Turner for providing coi1-1 seeds, J. Ecker for providing ein2-1 seeds, J. Zhou for providing att1 seeds, A. Yephremov for providing bdg and lcr seeds, F.M. Ausubel for providing the B. cinerea strain, and T. Mengiste for providing an A. brassicicola strain. We also thank the Arabidopsis Biological Resource Center at The Ohio State University for providing BAC clone F13F21.
LITERATURE CITED
Bisgrove SR, Simonich MT, Smith NM, Sattler A, Innes RW (
Chassot C, Métraux J-P (
Frye CA, Innes RW (
Klement Z (
Kunkel BN, Bent AF, Dahlbeck D, Innes RW, Staskawicz BJ (
Sieber P, Schorderet M, Ryser U, Buchala A, Kolattukudy P, Metraux JP, Nawrath C (
Thomma B, Eggermont K, Penninckx I, Mauch-Mani B, Vogelsang R, Cammue BPA, Broekaert WF (
Warren RF, Henk A, Mowery P, Holub E, Innes RW (
Wellesen K, Durst F, Pinot F, Benveniste I, Nettesheim K, Wisman E, Steiner-Lange S, Saedler H, Yephremov A (
Whalen MC, Innes RW, Bent AF, Staskawicz BJ (
Author notes
This work was supported by the National Institutes of Health (NIH; grant nos. R01 GM63761 and R01 GM046451 to R.W.I.). M.T.S. was supported by an NIH training grant in genetics (GM 07757).
Present address: State Key Laboratory of Plant Cell and Chromosome Engineering, Institute Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing 100101, China.
Present address: Linus Pauling Institute, Oregon State University, Corvallis, OR 97331–6512.
Corresponding author; e-mail dztang@genetics.ac.cn; fax 86–10–6484–7489.
The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Dingzhong Tang (dztang@genetics.ac.cn).
Open Access articles can be viewed online without a subscription.
