These authors contributed equally to this work.
Absence of the endo-β-1,4-glucanases Cel1 and Cel2 reduces susceptibility to Botrytis cinerea in tomato
Article first published online: 3 OCT 2007
The Plant Journal
Volume 52, Issue 6, pages 1027–1040, December 2007
How to Cite
Flors, V., Leyva, M. d. l. O., Vicedo, B., Finiti, I., Real, M. D., García-Agustín, P., Bennett, A. B. and González-Bosch, C. (2007), Absence of the endo-β-1,4-glucanases Cel1 and Cel2 reduces susceptibility to Botrytis cinerea in tomato. The Plant Journal, 52: 1027–1040. doi: 10.1111/j.1365-313X.2007.03299.x
- Issue published online: 8 OCT 2007
- Article first published online: 3 OCT 2007
- Received 21 June 2007; revised 31 July 2007; accepted 7 August 2007.
- Botrytis cinerea;
- defense responses;
- cell wall
Cel1 and Cel2 are members of the tomato (Solanum lycopersicum Mill) endo-β-1,4-glucanase (EGase) family that may play a role in fruit ripening and organ abscission. This work demonstrates that Cel1 protein is present in other vegetative tissues and accumulates during leaf development. We recently reported the downregulation of both the Cel1 mRNA and protein upon fungal infection, suggesting the involvement of EGases in plant–pathogen interactions. This hypothesis was confirmed by assessing the resistance to Botrytis cinerea infection of transgenic plants expressing both genes in an antisense orientation (Anti-Cel1, Anti-Cel2 and Anti-Cel1-Cel2). The Anti-Cel1-Cel2 plants showed enhanced resistance to this fungal necrotroph. Microscopical analysis of infected leaves revealed that tomato plants accumulated pathogen-inducible callose within the expanding lesion. Anti-Cel1-Cel2 plants presented a faster and enhanced callose accumulation against B. cinerea than wild-type plants. The inhibitor 2-deoxy-d-glucose, a callose synthesis inhibitor, showed a direct relationship between faster callose accumulation and enhanced resistance to B. cinerea. EGase activity appears to negatively modulate callose deposition. The absence of both EGase genes was associated with changes in the expression of the pathogen-related genes PR1 and LoxD. Interestingly, Anti-Cel1-Cel2 plants were more susceptible to Pseudomonas syringae, displaying severe disease symptoms and enhanced bacterial growth relative to wild-type plants. Analysis of the involvement of Cel1 and Cel2 in the susceptibility to B. cinerea in fruits was done with the ripening-impaired mutants Never ripe (Nr) and Ripening inhibitor (rin). The data reported in this work support the idea that enzymes involved in cell wall metabolism play a role in susceptibility to pathogens.
Tomato (Solanum lycopersicum Mill) β-1,4-endoglucanases (EGases) constitute a family of eight divergent members involved in different plant processes including fruit ripening, organ abscission and cell expansion and differentiation (Brummell et al., 1997; del Campillo and Bennett, 1996; Cataláet al., 1997; Lashbrook et al., 1994; Milligan and Gasser, 1995). These hydrolytic enzymes are likely to be involved in modifications to xyloglucan associated with the cell wall disassembly which takes place during these processes. Two members of the EGase gene family, Cel1 and Cel2, have been involved in fruit ripening and flower abscission. A correlation between accumulation of Cel1 and Cel2 mRNA and the progression of both processes was previously demonstrated (González-Bosch et al., 1996; Lashbrook et al., 1994). Specific antibodies against Cel1 were obtained and served to demonstrate the participation of this protein during flower abscission (González-Bosch et al., 1997). The Cel1 protein was also characterized in fruit and has several polypeptides (Real et al., 2004), possibly as a result of variable amounts of either glycosylation at the potential N-glycosylation sites reported in the sequence analysis (Lashbrook et al., 1994) or of other post-translational processes. The expression pattern of Cel1 during fruit ripening supported a possible role of this EGase in both the softening of pericarp tissue and the liquefaction of locules that takes place during ripening. We also demonstrated that the Cel1 protein and mRNA levels were down-regulated in pericarp after Botrytis cinerea infection (Real et al., 2004). This report indicates the differential expression of a EGase after fungal infection, and suggests a relationship between EGases, plant defense responses and fruit ripening.
Plants have evolved various defense mechanisms against pathogens. Apart from specific defense responses based on the so-called R-genes against certain strains of a pathogen, plants present a broad spectrum of defense responses which are pre-formed or can be induced locally or systemically by biotic or abiotic agents in nature (Durrant and Dong, 2004). The induced protection is mediated by multiple signal transduction pathways involving salicylic acid (SA), jasmonic acid (JA) and ethylene (ET); these can crosstalk with each other and lead to the expression of different responses including the hypersensitive response (HR), cell wall strengthening, the oxidative burst and the expression of various defense-related genes (Bostock, 2005). The cell wall not only plays a role as a constitutive barrier, but also as an active defense to stop pathogens. To prevent infection, attacked cells respond by a local reinforcement of the cell wall beneath the site of penetration attempt by forming a papilla. This process involves callose deposition in addition to the accumulation of H2O2 and phenolic compounds, as well as increased amounts of proteins and glycoproteins with hydrolytic and antifungal properties (Kruger et al., 2002). Callose, a sugar polymer that is composed of (1,3)-β-d-glucan, is deposited between the plasma membrane and the cell wall in close proximity to the invading pathogen as part of the HR to infection by fungi and viruses (Donofrio and Delaney, 2001; Ryals et al., 1996) and also in response to abiotic stresses (Currier, 1957; Currier and Webster, 1964; Mcnairn, 1972). Accumulation of callose in response to pathogenic attacks has been well documented in the interaction between barley (Hordeum vulgare) and the powdery mildew fungus Blumeria graminis f. sp. Hordei (Kruger et al., 2002). Recently, the accumulation of callose has been described as a major component of the defense mechanisms against necrotrophs in Arabidopsis (Flors et al., 2005; Ton and Mauch-Mani, 2004; Zimmerli et al., 2000, 2001). Callose deposition in tomato plants was reported after infection with cucumber mosaic virus (Xu et al., 2003). Several studies support the hypothesis that the cell wall must contain inherent signaling properties, essential for protection against pathogens as well as for coordinated growth and development. There is plenty of evidence to show that the plant cell wall is involved in multiple signaling pathways. Plant cells can perceive changes in the wall composition, and cells respond by activating signaling networks, thus providing a sensing mechanism through which several responses can be properly coordinated or altered (Pilling and Hofte, 2003; Vorwerk et al., 2004). Three major groups of cell wall fragments appear to play important roles in the initiation of a defense response: fungal chitin, fungal glucans and plant oligogalacturonides (OGAs). Fragments of OGAs released from pectin by the degrading activity of pathogen-secreted or endogenous plant polygalacturonases (PGs) and pectate lyases are involved in a variety of signal transduction pathways that regulate normal plant growth and development, and activate defense responses (Pilling and Hofte, 2003; Vorwerk et al., 2004). Several reports indicate that the induction of some defense genes by OGAs is mediated through JA (Doares et al., 1995), and that the application of OGAs to tobacco suspension-cultured cells also induces accumulation of SA (Klarzynski et al., 2000). Thus, OGAs apparently activate two of the major disease-related signal systems.
The study of cell wall hydrolytic enzymes in plant–pathogen interactions has focused on activities that degrade polysaccharides in the cell wall of the invading pathogen. These include β-1,3-glucanases and chitinases, usually considered as pathogenesis-related (PR) proteins because of their induction upon infection (van Loon et al., 2006). However, the role in the pathogen resistance played by cell wall-degrading enzymes associated with plant growth and development has received little attention. Previous reports suggest that cell wall-degrading enzymes are required for susceptibility. Mutations in the Arabidopsis PMR6 gene, encoding a pectate lyase-like protein, confer strong resistance to powdery mildew (Erysiphe cichoracearum), hence PMR6 appears to be required for susceptibility to powdery mildew (Vogel et al., 2002). The pmr6-mediated resistance is mainly based on changes in the composition of the cell wall and does not require the activation of known host defense pathways. Thus, pmr6 resistance represents a novel form of disease resistance based on the loss of a gene which is required during a compatible interaction. It has recently been demonstrated that plants in which pectin methylesterase (PME) inhibitors are overexpressed are more resistant to the necrotrophic fungus B. cinerea; this is related to its impaired ability to grow on methylesterified pectins (Lionetti et al., 2007). The role of PG in the development of susceptibility to post-harvest pathogens in ripening tomato has also been studied. In this case, some reports suggested that fruits with reduced PG were more resistant to Geotrichum candidum and Rhizopus stolonifer (Kramer et al., 1992). However, further reports indicated that transgenic lines deficient in PG showed the same susceptibility to Colletotrichum gloeosporoides as wild-type fruit (Cooper et al., 1998). It is clear that while there is a reasonable knowledge of plant genes involved in resistance little is known about those required for susceptibility. Our objective was to investigate the role of tomato EGases in plant–pathogen interactions. The demonstration of Cel1 protein accumulation in fully expanded leaves and the establishment of an infection system based on pathogen inoculation of the whole plant allowed us to conduct studies in both tomato plants and fruit. With this aim, transgenic plants and fruits, with constitutively reduced levels of Cel1 and/or Cel2 expression (Brummell and Harpster, 2001; Brummell et al., 1999; Lashbrook et al., 1998) were studied following B. cinerea infection. Studies were also performed with fruits from the tomato ripening-impaired mutants Never ripe (Nr) and Ripening inhibitor (rin), in which a reduced expression of Cel1 and Cel2 mRNAs had been previously demonstrated (González-Bosch et al., 1996). Fruits of rin mutants, in which Cel2 had been expressed (ABB, unpublished), were also analyzed. Microscopical studies of infected leaves and the analysis of marker genes for the main signal transduction pathways were also performed. In addition, transgenic antisense plants were challenged with another pathogen Pseudomonas syringae. Results are discussed in terms of both the requirement of EGases for susceptibility to B. cinerea and the possible link between reduction of EGases and plant responses against pathogens.
Cel1 is expressed in different plant tissues but is absent in transgenic antisense plants
Both Cel1 and Cel2 mRNA transcripts were previously reported to be present in a wide variety of tomato plant tissues, albeit at different levels (Lashbrook et al., 1994). In fact, only ripening tomato fruit and flower abscission zones had a high level of total EGase mRNA. Specific polyclonal antibodies were raised against Cel1 to allow the characterization of the Cel1 protein throughout flower abscission and fruit ripening (González-Bosch et al., 1997; Real et al., 2004). Here, we analyzed the presence of the Cel1 protein in different parts of tomato plants. Despite the low expression of the Cel1 gene previously reported in some tissues, we demonstrated that the protein was present in leaves, shoots and petioles, but absent in roots (Figure 1a). We also assessed whether the Cel1 protein accumulated during leaf development (Figure 1b). In this tissue, a pattern of bands similar to those found in fruit and abscission zones was obtained by using specific polyclonal antibodies against tomato Cel1 (González-Bosch et al., 1997), supporting a variable level of glycosylation at the potential sites reported by the sequence analysis or post-translational modification (Lashbrook et al., 1994). Cel1 is absent in immature leaves, but accumulates in fully expanded leaves and decreases in abundance in senescent leaves (Figure 1b). The presence of the Cel1 protein in this tissue and the accumulation pattern observed point to a possible role of this EGase in both leaf development and processes in which the cell wall is involved, such as plant–pathogen interaction.
The absence of the Cel1 protein was confirmed in leaves and fruit from Anti-Cel1 and Anti-Cel1-Cel2 plants (Figure 1c). We highlight the fact that this protein was almost undetectable in pericarp tissues of wild-type plants but present at considerable levels in locular tissues. The small amount of Cel1 protein in the fruit pericarp of the T5 cultivar made it difficult to assess the role of Cel1 in processes associated with this tissue since the transgenic lines were developed in this cultivar. Cel2 antisense plants accumulated the Cel1 protein in leaves and fruit locules at levels similar to those observed in wild-type plants (Figure 2), which suggests that expression of Cel1 does not compensate for reduced levels of Cel2.
Absence of EGases increases resistance against B. cinerea
In a search to establish the possible involvement of cell wall EGases in plant–pathogen interactions, laboratory-controlled infections with the fungal necrotroph B. cinerea were performed. It was observed that detached mature leaves were more susceptible to B. cinerea than young leaves (Figure 2). This observation is consistent with a possible contribution of this EGase to the basal susceptibility to B. cinerea, as suggested by the accumulation of Cel1 protein throughout leaf development (Figure 1b). To analyze this possibility, a whole-plant inoculation protocol was set up to obtain a more reliable and reproducible system for infection studies (see Experimental procedures for details) than those based on detached leaves. Four-week-old plants expressing the EGase genes in an antisense orientation were inoculated with the pathogen and compared with wild-type plants. Figure 3(a) shows how wild-type plants (cv. T5) were susceptible to B. cinerea infection. Necroses upon fungal infection were observed 48 h after challenges. Disease symptoms progressively increased at 72 and 96 h, and the leaf tissue was completely macerated 6 days after inoculation (data not shown). Anti-Cel1 and Anti-Cel2 plants presented a wild-type phenotype. However, Anti-Cel1-Cel2 plants revealed an enhanced resistance compared with wild-type plants, as indicated by the reduced necrosis. Most of the Anti-Cel1-Cel2 inoculated leaves showed necrotic areas with more defined limits than wild-type ones (Figure 3b). Abolishing a single EGase activity may not be enough to modify plant basal resistance, although blocking both EGases led to an enhanced resistant phenotype. These results indicate that EGases could constitute susceptibility factors for pathogen infection because their reduction increased resistance of the plant to B. cinerea.
Botrytis cinerea resistance induced by reduction of EGases is related to faster and enhanced callose deposition
To gain more insight into the nature of the induced resistance against B. cinerea of plants with reduced levels of EGases, cytological observations were carried out at the sites of infection. Wild-type plants accumulated increasing amounts of callose in a time-dependent manner after inoculation with B. cinerea (Figure 4a), while no pathogen-inducible callose was present in uninoculated leaves (data not shown). Similar to other compatible interactions, this defense reaction was not fast enough to stop the pathogen. Interestingly, the reduction of both EGases produced a significant increase in callose deposition in the epidermal cell layer in comparison with control plants after infection with B. cinerea (Figure 4a). Anti-Cel1-Cel2 plants accumulated more callose at the limits of the infected area (Figure 4b) and at distal parts from the directly infected tissues (not shown). No callose deposition was observed in uninoculated Anti-Cel1-Cel2 plants (not shown). Anti-Cel1 and Anti-Cel2 plants did not display significant differences in callose deposition in comparison with wild-type plants after perception of the pathogen (Figure 4a). Therefore, it seems that the absence of both EGases primes the accumulation of callose that may be associated with pathogen perception or might be a result of some signal that originated during the infection process. Previous reports have shown callose accumulation as a defense mechanism acting in tomato against CMV (Xu et al., 2003) and also in Arabidopsis infected with the necrotroph Alternaria brassicicola (Ton and Mauch-Mani, 2004).
To confirm whether callose deposition was related to the enhanced resistance of Anti-Cel1-Cel2 plants against B. cinerea, 2-deoxy-d-glucose (2-DDG), an inhibitor of callose synthesis, was used (Jaffe and Leopold, 1984). Wild-type and Anti-Cel1-Cel2 plants were treated by infiltrating leaves with 2-DDG and inoculating them in the infiltrated area 24 h later with B. cinerea spores. The rate of infection was estimated by measuring the average lesion diameters (Figure 5a). Control plants showed enhanced susceptibility to B. cinerea, and it was confirmed that Anti-Cel1-Cel2 plants had lost their enhanced basal resistance to the necrotroph in the presence of 2-DDG. Local callose accumulation was visualized microscopically 96 h after fungal inoculation (Figure 5b). After treatment, pathogen-induced callose production was severely reduced in both control plants and Anti-Cel1-Cel2 plants. Normal callose production was observed outside the infiltration site, as expected for local treatment.
SA- and JA-dependent defenses are induced by reduction of EGase
To establish whether the enhanced resistance of transgenic plants led to an induction of SA- and/or JA-mediated defenses, the expression of PR1 and LoxD marker genes was analyzed by RT-quantitative (q)PCR. Both genes have been previously reported to be pathogen-inducible in tomato (Benito et al., 1998; Cohn and Martin, 2005). As shown in Figure 6(a), PR1 expression was induced in wild-type plants in response to B. cinerea infection, which is consistent with reports that the SA-dependent pathway is effective against B. cinerea in tomato (Achuo et al., 2004). The antisense expression of Cel1 and/or Cel2 enhanced the accumulation of PR1 transcript compared with wild-type plants after fungal challenge. LoxD was also induced in wild-type plants upon infection with B. cinerea, supporting the involvement of JA-inducible defenses in this plant–pathogen interaction, as previously reported (Diaz et al., 2002). LoxD expression was also enhanced in all the antisense plants in response to fungal infection (Figure 6b). These results suggest that EGase activity influences SA-mediated and JA-mediated defense responses in tomato.
The absence of EGases increases susceptibility to P. syringae
To establish whether the resistance showed by the Anti-Cel1-Cel2 plants was specific for B. cinerea, these plants were challenged with another pathogen, P. syringae. Anti-Cel1-Cel2 plants inoculated with the strain Pst DC3000 consistently showed classic symptoms of susceptibility, such as chlorosis, water-soaked lesions and necrotic pits. These symptoms were more severe in antisense plants than in wild-type plants. Furthermore, there was also a significant increase in bacterial growth in the Anti-Cel1-Cel2 plants 72 h after infection (Figure 7). These results indicate that the absence of both EGases increases susceptibility of the plant to P. syringae. Therefore the absence of these EGases influences the responses of the plant to different pathogens, which makes tomato plants more susceptible to the bacterial pathogen Pst DC3000 but more resistant to the necrotrophic fungus B. cinerea.
Analysis of fungal infection in transgenic tomato fruits
The response to B. cinerea was studied in fruits from transgenic antisense plants to observe the effects caused by the absence of Cel1 and Cel2. In this case, conidial suspensions were inoculated onto the wounded outer peel of tomato fruits at different ripening stages, as previously described (Real et al., 2004). The diameter of the necrosis was measured 96 h after inoculation, and the measurements found were 0.92 ± 0.04 cm in mature green fruits (MG4) and 1.19 ± 0.06 cm in ripe fruits (red). No significant differences were observed in the necrotic area between transgenic and wild-type fruits. The extremely low levels of the Cel1 protein detected in the pericarp of wild-type fruits made it difficult to observe significant differences from transgenic fruits in this tissue. In addition, the requirement of fruit wounding for necrotroph inoculation could mask early defenses at the penetration sites. To confirm whether the induced injury masked the role of EGases, a less aggressive method of fruit inoculation was set up consisting of the conidial pulverization of intact fruits. Under these inoculation conditions, we could observe differences in susceptibility among varieties not detected when inoculations were performed by wounding, thus supporting the benefits of spray inoculation in these studies (Figure 8). Indeed when spores were inoculated in wounded fruits, they all showed disease symptoms with no significant differences in necrosis diameter, irrespective of either the cultivar or the ripening stages of fruits. However, when spores were spray inoculated, significant differences in the percentage of infected fruits were obtained among varieties at all the ripening stages studied. Transgenic fruits were then inoculated by this system, but no significant differences were observed in the number of infected fruits in comparison with wild-type fruits (data not shown).
Analysis of fungal infection in fruits from tomato ripening mutants
To further assess the role of EGases in the response of tomato fruits to fungal pathogens, the ripening mutants Nr and rin were challenged by B. cinerea. These mutants contained reduced levels of Cel1 and Cel2 mRNAs in pericarp and locules (González-Bosch et al., 1996). In fact, Cel2 mRNA was completely absent in both tissues in rin fruits. For this reason, the Cel2 gene was expressed in this mutant under the control of the fruit-specific E8 promoter (ABB, unpublished) to test whether the constitutive absence of Cel2 influenced the mutant phenotype and to also study the role of this EGase in fruit. An accumulation of Cel2 mRNA was confirmed in the rin-Cel2 fruits by RT-qPCR (data not shown). The Cel2 protein could not be analyzed because specific Cel2 antibodies were not yet available. Transgenic rin fruits expressing Cel2 (rin-Cel2) are still deficient in ripening, and show the characteristic yellow color of mature rin fruits. However, the presence of Cel2 accelerates the ripening process, particularly the rate of fruit softening (data not shown).
Infection experiments were performed by either inoculating B. cinerea conidia on wounded tomatoes or alternatively spraying them on the fruit surface, as previously described. When inoculation was carried out by wounding, fruits from Nr mutants at different ripening stages showed a similar susceptibility to their corresponding wild-type counterparts (data not shown). Similar results were obtained when conidia were sprayed onto Nr fruits (data not shown). When rin mutants were challenged with B. cinerea, fruits were slightly more resistant to infection than wild-type fruits (cv. Ailsa Craig) in both infection systems (wounding and spray) (Figure 9). However, the progression of the B. cinerea infection in the rin-Cel2 fruits was significantly higher than in rin mutants among wild-type fruits (Figure 9).
The presence of the Cel1 protein in all these mutants was analyzed by using Cel1 antibodies. As shown in Figure 10, the absence of the Cel1 protein throughout ripening was confirmed in Nr, as was a delay in its accumulation in the rin and rin-Cel2 fruits, where the Cel1 protein accumulated at very late stages, corresponding to an overripe stage of the respective controls. These results confirm the low levels of Cel1 in these mutants, but point to a complex regulation of this EGase, which permits a final accumulation of the polypeptides at very late ripening stages.
We have shown that the absence of the EGases Cel1 and Cel2 makes tomato plants more resistant to the necrotrophic fungus B. cinerea. Previous results showed that the pattern of gene expression of these EGases suggested a role in fruit ripening and flower abscission (Lashbrook et al., 1994). The study of the Cel1 protein supported its involvement in both processes (González-Bosch et al., 1997; Real et al., 2004). Here we demonstrated that Cel1 is also present in other vegetative tissues, such as leaves and shoots. Cel1 antiserum identified a pattern of several polypeptides, which is consistent with that reported for flower abscission zones and fruit (González-Bosch et al., 1997; Real et al., 2004). Interestingly, accumulation of Cel1 protein was associated with leaf development, and the highest level was noted in fully expanded leaves. Recently, we reported the differential expression of Cel1 in fruits infected by fungal necrotrophs (Real et al., 2004). These results supported an interesting relationship between EGases, plant defenses and fruit ripening. The presence of the Cel1 protein and its pattern of accumulation in leaves allowed the study of the role played by this EGase in the basal resistance against pathogens in whole plants. Very little information is available about the role of cell wall hydrolytic enzymes in plant–pathogen interactions, except for those usually considered as pathogenesis-related (PR) proteins, such as β-1,3-glucanases and chitinases (van Loon et al., 2006). The PMR6 gene, which encodes a pectate lyase-like protein, has been reported to be necessary for susceptibility to powdery mildew (E. cichoracearum) in Arabidopsis because mutations in this gene confer a strong resistance to this pathogen (Vogel et al., 2002). Although it is not yet known why the loss of a plant pectate lyase causes resistance, the observation is consistent with the idea that polysaccharide composition is a determinant of disease interactions. In this work we have studied the involvement of the tomato EGases Cel1 and Cel2 in a plant–pathogen interaction. We have analyzed tomato transgenic plants with constitutively reduced levels of these EGases upon fungal infection. The antisense plants for both genes (Anti-Cel1-Cel2) were significantly more resistant to B. cinerea infection, showing a phenotype which consisted of a more restricted necrotic area at the infection site. The enhanced resistance of the double antisense plants indicates that Cel1 and Cel2 may play a role in the susceptibility of tomato to B. cinerea. In this work, the study of the cell wall structure within the expanding lesion and the surrounding healthy tissue has shown that tomato plants accumulated pathogen-inducible callose upon fungal infection. These results suggested that callose deposition constitutes a defense mechanism against the necrotroph B. cinerea in tomato. Iglesias and Meins (2000) proved that the movement of plant viruses is delayed in a β-1,3-glucanase-deficient mutant, showing an enhanced resistance related to augmented deposition of callose. Inhibition of callose synthesis by infiltrating 2-DDG, a callose synthesis inhibitor (Flors et al., 2005; Ton and Mauch-Mani, 2004), reversed the enhanced resistance of the Anti-Cel1-Cel2 plants to the wild-type levels. This supports the involvement of this mechanism in inducible defenses against necrotrophs and the relationship between EGase activity and protection upon fungal infection.
To establish whether the increased resistance phenotype observed in the Anti-Cel1-Cel2 transgenic plants led to an induction of SA- and/or JA-mediated defenses, we analyzed the expression of PR1 and LoxD marker genes for both pathways. There are several reports showing that the SA-dependent pathway is effective against B. cinerea in tomato (Achuo et al., 2004) as opposed to the observation that SA signaling does not appear to play a major role in the resistance to B. cinerea in Arabidopsis (Glazebrook, 2005). The expression of the PR1 gene increased in wild-type plants in response to B. cinerea infection, thus supporting the involvement of SA-mediated responses in this plant–pathogen interaction. The absence of EGases in the antisense plants led to an additional increase of the PR1 transcript level. Analysis of LoxD expression showed that this gene was induced in tomato plants infected by B. cinerea, supporting the involvement of JA-inducible defenses in this interaction (Diaz et al., 2002). These authors showed that JA acted in the resistance of tomato to B. cinerea based on the increased susceptibility shown by the JA-deficient mutant def1. Contradictory results were obtained by Audenaert et al. (2002), who reported that the mutant def1 did not show increased susceptibility to B. cinerea. The antisense expression of Cel1 and/or Cel2 genes caused increased expression of LoxD in response to fungal infection. The fact that the absence of Cel1 and Cel2 is accompanied by increased expression of genes involved in the main defense signal pathways provides more insight into their roles in plant responses.
The study of how EGases are involved in plant–pathogen interactions was extended to fruits challenged with B. cinerea. In this case, no significant differences in susceptibility were observed between transgenic and wild-type fruits. This result could reflect inherent differences in both tissue composition and response mechanisms against B. cinerea in tomato fruits in comparison with leaves. A less aggressive spray inoculation method was set up to avoid injuries that could mask changes in susceptibility. This infection protocol allowed differences in susceptibility among tomato cultivars to be seen, while the same rate of infection was observed for transgenic anti-Cel1-Cel2 fruits in relation to wild-type fruits. These observations indicated that transgenic fruits do not enable the effect of EGase reduction on pathogen resistance to be studied. The ripening-impaired mutants Nr and rin were used in an attempt to further assess the role of EGases in the tomato fruit–pathogen interactions. Previous analyses of Cel1 and Cel2 mRNA accumulation in both mutants provided evidence that they accumulated at very low levels and subjected to a different regulatory control (González-Bosch et al., 1996). The most striking feature of that study was that Cel2 mRNA was absent in rin fruits and was also insensitive to ethylene in this genetic background (González-Bosch et al., 1996), showing many characteristics similar to those of the PG gene (DellaPenna et al., 1989; Yen et al., 1995). The absence of the Cel2 transcript in the rin mutant suggested a possible correlation between the activity of this EGase and the ripening deficiency of rin fruits. This idea prompted the testing of transgenic rin fruits expressing a Cel2 gene (ABB, unpublished), referred to in this work as rin-Cel2. These transgenic fruits were still deficient in ripening, although the presence of Cel2 increased the rate of fruit softening. An analysis of the Cel1 protein demonstrated a delay in its accumulation in rin and Nr fruits throughout the period corresponding to the ripening in wild-type fruits. A similar pattern to that observed in rin fruits was obtained in rin-Cel2 fruits, indicating that the expression of Cel2 in this mutant did not affect accumulation of Cel1 protein. Interestingly, the analysis of fruit mutants left to ripen in plants for long periods of time revealed that the Cel1 protein finally accumulated in all cases. These results demonstrated that at least the Cel1 protein escaped from the down-regulation in ripe mutant fruits, and provided additional data regarding the complex control of EGase expression in these ripening mutants. Fruits from the Nr and rin mutants were challenged by B. cinerea to study fungal susceptibility. Nr fruits showed a similar susceptibility to fungal infection as their corresponding wild-type counterparts (cv. Pearson) as previously reported for Nr plants (Diaz et al., 2002). rin fruits challenged by B. cinerea were slightly more resistant to infection than wild-type fruits (cv. Ailsa Craig), and this can be interpreted in terms of a minor contribution of this EGase to the susceptibility to B. cinerea in fruit. However, the expression of Cel2 in this genetic background made the fruits become more sensitive to the pathogen. The mechanisms by which tomato fruits respond to fungal infection are not well established and could be different from those acting in leaves. A constitutive reduction of one or both EGases does not seem sufficient to affect the fruit basal susceptibility, but the presence of Cel2 in fruits lacking this EGase facilitates the expansion of B. cinerea. Apparently, the pathogen takes advantage of the hydrolytic activity of Cel2, which could be interpreted in terms of an increase in fruit softening that favors the spread of the pathogen. The possibility that this EGase contributes to fruit basal susceptibility in a different way to which it does in leaves cannot be ruled out.
Collectively, the results obtained from the studies carried out in plants and fruits enable us to suggest that a connection exists between EGases and susceptibility to the fungal necrotroph B. cinerea in tomato. It has been also demonstrated that the Anti-Cel1-Cel2 plants display severe disease symptoms and enhanced bacterial growth with regard to wild-type plants when inoculated with P. syringae. This result indicates that the absence of both EGases influences the responses of the plant to different pathogens and supports a role for these cell wall hydrolytic enzymes in plant susceptibility. Further studies should be done to verify whether the absence of EGases produces changes in both cell wall composition and gene regulation in such a way that fungal progress is slowed but bacterial spread is facilitated. Different sets of defense responses are known to be activated against necrotrophic and biotrophic pathogens, even though the relationships between different signaling pathways are complex and not yet clear (Glazebrook, 2005). On the other hand, little is known about the biochemical properties and substrate specificities of EGases except that potential wall substrates include cellulose and xyloglucan (Cosgrove, 2005). Endo-β-1,4-glucanases have been related to hemicellulose degradation during vegetative growth and fruit ripening (Hatfield and Nevins, 1986; Hayashi et al., 1984; Libertini et al., 2004; Rose and Bennett, 1999). However, two EGases from poplar (PopCel1 and PopCel2) were examined and indirect evidence that they digest non-crystalline regions of cellulose was presented (Ohmiya et al., 2000). Based on the apparently broad spectrum of substrates for EGases, a possible explanation for the increased accumulation of pathogen-induced callose in the Anti-Cel1-Cel2 plants upon B. cinerea infection is that callose could also be an EGase substrate owing to the unspecific O-glycosyl activity of these enzymes. This enzymatic activity could prevent the accumulation of callose in the absence of pathogen challenges. A speculative but particularly attractive possibility is that the absence of the hydrolytic activity of EGases might be sensed by plant cells as a signal of the alteration of the regular structure and functioning of the cell wall to activate a battery of defenses in order to challenge a potential stress situation. If this were the case, products from EGase activity could act as signals of normal cell wall metabolism to prevent the induction of defense mechanisms. The involvement of the cell wall in signaling has received considerable backing in recent years. Several publications demonstrate the involvement of different cell wall components and enzymatic activities in diverse plant–pathogen interactions. Mutations in cellulose synthases have been reported to confer enhanced resistance to the soil-borne bacterium Ralstonia solanacearum and the necrotrophic fungus Plectosphaerella cucumerina in Arabidopsis (Hernández-Blanco et al., 2007). These results also indicate that an alteration of the secondary cell wall structure by inhibition of cellulose synthesis leads to a specific activation of novel defense pathways irrespectively of SA, ethylene and JA signaling. The overexpression of pectin methylesterase (PME) inhibitors has also been seen to result in a decrease of PME activity in transgenic plants, and the degree of pectin methylesterification increased. Transformed plants were more resistant to the necrotrophic fungus B. cinerea, which was related to its impaired ability to grow on methylesterified pectins (Lionetti et al., 2007). The Arabidopsis thaliana mutant sma4 displays severe disease symptoms when inoculated with avirulent strains of P. syringae pv. tomato, but is highly resistant to B. cinerea. SMA4 encodes LACS2, a member of the long chain acyl-CoA synthetases, that is involved in cutin biosynthesis. The authors conclude that plant cutin, or the cuticle structure, may play a key role in tolerance to biotic and abiotic stress and in the pathogenesis of B. cinerea (Tang et al., 2007). The botrytis-resistant 1 (bre1) mutant of Arabidopsis reveals that a permeable cuticle does not facilitate the entry of fungal pathogens in general, but surprisingly causes an arrest of invasion by Botrytis. BRE1 was identified as being long-chain acyl-CoA synthetase2 (LACS2; Bessire et al., 2007).
Plant cell walls are highly complex in structure and in composition. Albersheim and colleagues suggested that some of the structural complexity could represent latent signal molecules involved in defense rather than structures required for the mechanical function of the wall (Vorwerk et al., 2004). The possibility that EGase activity is part of the complex web of cell wall signaling and metabolism operating in plant–pathogen interactions opens an interesting field of research.
Different tomato (S. lycopersicum) genotypes were used in our studies. Tomato plants of cv. T5, either wild type or transformed with antisense genes for Cel1 (Lashbrook et al., 1998), Cel2 (Brummell et al., 1999) and both Cel1-Cel2 (Brummell and Harpster, 2001). The antisense Cel1-Cel2 tomato line (Anti-Cel1-Cel2) was generated by crossing plants previously characterized with 95% reduction in expression of either Cel1 (Lashbrook et al., 1998) or Cel2 (Brummell et al., 1999). Fruits resulting from this cross-pollination of transgenic lines were allowed to develop and 50 seeds were germinated. After screening these individuals by PCR, five lines were identified that had both antisense Cel1 and antisense Cel2 transgenes. These five plants were grown to maturity, 100 seeds were collected and assayed by genomic gel blot analysis to identify lines that were homozygous for both antisense Cel1 and Cel2 transgenes. A single Anti-Cel1-Cel2 line was selected for further characterization. These plants showed the wild-type phenotype except for a slight delay in plant development.
For in planta studies, seeds were germinated in vermiculite. After 4 weeks, seedlings were transferred to hydroponic conditions in tanks containing Hoagland solution. After 4 days, plants were inoculated with pathogen. Plants were grown at 24°C day/18°C night, with 16-h light/8-h dark and a relative humidity (RH) of 60%.
For fruit studies, fruits were harvested from greenhouse-grown plants and staged by color and ethylene production into defined developmental and ripening stages as follows: MG1 (MG, mature green), green fruit with solid locules but fully developed seeds; MG4, all locules liquefied with slight pink coloration, seeds fully developed, no exterior color; BR (breaker), slight red color visible on the exterior of fruit at blossom end [ethylene 4–7 nl (g FW)−1 h−1]; TU (turning) fruit 10–30% light red [ethylene 7–10 nl (g FW)−1 h−1]; PK (pink) fruit 30–70% light red [ethylene 8–10 nl (g FW)−1 h−1]; LR (light red) fruit 100% light red [ethylene 8–10 nl (g FW)−1 h−1]; R (red ripe) fruit red and table ripe [ethylene 11–14 nl (g FW)−1 h−1]; and OR (over-ripe) fruit dark red and soft [ethylene 2–6 nl (g FW)−1 h−1] (FW is fresh weight). Fruits were collected either for immediate freezing or for subsequent pathogen inoculation.
Tomato plants of cv. Ailsa Craig, either wild type or nearly isogenic for the rin ripening mutation and of cv. Pearson, either wild-type or nearly isogenic for the Nr ripening mutation, were used for studies in fruits. Flowers from the second to the fifth bunch were tagged as the swelling ovary reached 2–3 mm in diameter, and fruits were collected 40 days later (equivalent to the late mature green stage), 47 days later (equivalent to the pink stage) and 51 days later (equivalent to the red ripe stage). rin and Nr fruit at day 40 after anthesis were almost identical in size, color and morphology, and with a similar locular development of their corresponding wild-type counterparts, indicating that their development prior to ripening was similar and that the time selected represented a comparable physiological stage. Five fruits per treatment time were processed to separate locules and pericarp, and were immediately frozen, and a further five fruits were pathogen inoculated.
Microbial strains and growth conditions
The fungus used in this study was B. cinerea CECT2100 (Spanish collection of type cultures). It was routinely cultured on potato dextrose agar (PDA) (Difco; http://www.bd.com) at 24°C. Botrytis cinerea spores were collected from 10- to 15-day-old cultures with sterile water containing 0.01% (v/v) Tween-20, which was then filtered, quantified with a hemacytometer, and adjusted to an appropriate concentration.
Pseudomonas syringae Pst DC3000 was grown in a low-salt liquid Luria–Bertani (LB) medium (10 g l−1 tryptone, 5 g l−1 yeast extract and 5 g l−1 NaCl, pH 7.0) at 28°C. Rifampicin was added to LB at 50 μg ml−1. Bacterial suspensions were adjusted to 5 × 105 colony-forming units (CFU) ml−1 (OD600 = 0.001) in sterile distilled water containing 0.015% of the surfactant Silwet L-77 (Osi Specialties; http://www.ge.com), as previously described (Katagiri et al., 2002).
Botrytis cinerea inoculation on tomato plants and microscopic analysis
Conidia collected from 10- to 15-day-old PDA plates supplemented with 40 mg ml−1 of tomato leaves were maintained in Gambor’s B5 medium (Duchefa, http://www.duchefa.com/), supplemented with 10 mm sucrose and 10 mm KH2PO4 for 2 h in the dark with no shaking.
For inoculation of detached leaves, young and mature leaves were collected from 12-week-old tomato plants and placed in 5-cm Petri plates to allow the petiole to be in contact with water. Leaves were challenged by applying 5 μl droplets of a 1 × 106 spores ml−1 conidia suspension and placed in another 15-cm Petri plate, where high humidity was maintained. After challenge inoculation, plates were maintained at 20°C and a high RH. Fungal hyphae grew concentrically from the site of inoculation, resulting in visible necrosis 48 h after inoculation. After 72 h, the diameter of the necrotic area was measured.
For plant inoculation, 4-week-old plants were challenged by applying 5 μl droplets of 1 × 106 spores ml−1. After challenge inoculation, plants were maintained at a RH of 100%. Fungal hyphae grew concentrically from the site of inoculation, resulting in visible necrosis 48 h after inoculation. Disease symptoms were assessed by determining the average lesion diameter in 15 plants per genotype.
Callose deposition was determined in control and infected leaves at different time-points after inoculation by staining with calcofluor/aniline blue, and further analysis and quantification with epifluorescence microscopy were done, as described in Ton and Mauch-Mani (2004). To determine the effect of the callose inhibitor 2-DDG on resistance against B. cinerea, leaves of 4-week-old wild-type and antisense cultured plants were infiltrated with 2-DDG (1.2 mm) using a syringe without a needle, and were challenge inoculated 24 h later. Disease symptoms were assessed 96 h after the inoculation and 15 leaves were collected for callose staining.
Botrytis cinerea inoculation on tomato fruits
Two different methods were used – wounding and spray.
Inoculation by wounding.
Conidia of B. cinerea were collected from 10-day-old cultures in PDA supplemented with tomato leaf extract, and were filtered and quantified as previously described (Real et al., 2004). Freshly collected tomato fruits were surface-sterilized by 10-min incubation in 10% commercial bleach solution and subsequently washed in water three times. Fruits were air dried and wounded by making four punctures along the equatorial line. Control fruits were inoculated with 5 μl of sterilized distilled water in each wound (control fruits) or with 5 μl of a suspension of 4 × 106 conidia ml−1 (infected fruits). Inoculated fruits and the corresponding controls were maintained at 20°C and at a RH of 85% for 96 h and scored for symptoms.
Fruits were harvested and classified as above and sprayed three times with a suspension of 5 × 105 conidia ml−1. Control fruits were sprayed with sterilized water. Inoculated fruits and the corresponding controls were maintained at 20°C and at a RH of 85% for 1–3 weeks, depending on the cultivar, and scored for symptoms.
Pseudomonas syringae inoculation in tomato plants
Four-week-old tomato plants were dipping-inoculated with bacterial suspensions as previously described (Katagiri et al., 2002). Bacterial growth was monitored within leaf tissue by grinding 1 g of leaf tissue for plant and plating dilutions of the ground material on LB media with the appropriate antibiotics. At least three samples were taken for each treatment over a 3-day period. Each experiment was independently conducted at least three times.
Protein extraction and Western blotting
Soluble proteins from different tomato tissues were extracted by using the protocol previously described for flower abscission zones (González-Bosch et al., 1997) with minor modifications. Tissues were ground to a powder in liquid nitrogen with a mortar and pestle, and proteins were extracted with a buffer (2.2 ml g−1 FW) containing 20 mm 2-amino-2-hydroxymethyl)-1,3-propanediol (TRIS)–HCl (pH 8.0), 3 mm EDTA, 0.5 m NaCl. The mixture was homogenized with cold buffer for 1 min at high speed with a tissue homogenizer and then centrifuged at 10 000g for 10 min. The supernatant was filtered through GFA (Millipore, http://www.millipore.com/) filters and then precipitated by adding cold acetone to a final concentration of 80% (v/v). After incubation at 20°C for 16 h, the precipitate was collected by centrifugation at 15 000g for 15 min and resuspended in 20 mm TRIS–HCl (pH 8.0), 3 mm EDTA, 250 mm sucrose, 2 mm Na2S2O5. Protein content was measured in extracts by the Bradford method (Bradford, 1976) with standard curves prepared using bovine serum albumin (BSA) (Sigma, http://www.sigmaaldrich.com/).
Protein extracts were separated by SDS–PAGE (Laemmli, 1970) and transferred to nitrocellulose by electroblotting in 48 mm TRIS, 39 mm glycine, 0.03% SDS, 20% methanol. After transfer, the nitrocellulose was saturated with 3% BSA for 1 h at room temperature, and then incubated overnight at room temperature with antiserum diluted in 150 mm NaCl, 10 mm TRIS–HCl (pH 8.0) and 0.1% Tween-20. Antigen–antibody complexes were detected by alkaline phosphatase-conjugated goat anti-rabbit IgG (Bio-Rad, http://www.bio-rad.com/) diluted in the same buffer as the antiserum. The blot was washed in 150 mm NaCl, 10 mm TRIS–HCl (pH 8.0) and 0.1% Tween-20, and color developed using 5-bromo-4-chloro-3-indoyl phosphate p-toluidine salt and p-nitroblue tetrazolium chloride (Sigma). Pre-immune rabbit serum was used as a control.
RT-qPCR analysis of transcripts
Ribonucleic acid was extracted from tomato leaves with Total Quick RNA Cells and Tissues Kit (Talent.; http://www.spin.it/talent) at 0, 24, 48 and 72 h after inoculation. Leaf tissue was collected from 15 wild-type and antisense plants (grown in parallel). For RT-qPCR experiments, 1.5 μg of total RNA was digested with 1 unit of DNase (RNase free) in 1 μl of DNase buffer and Milli-Q water up to 10 μl (Promega, RQ1 RNase-Free DNase, http://www.promega.com/) and incubated for 30 min at 37°C. After incubation, 1 μl of RQ1 DNase stop buffer was added and incubated again at 65°C for 10 min to inactivate DNase. The RT reaction was performed by adding 2μl of RT buffer, 2 μl of 5 mm dNTP, 2 μl of 10 μm Oligo(dT)15 primer (Promega, Oligo(dT)15 Primer), 1 μl of 10 U/μl RNase inhibitor (Promega Rnasin RNase inhibitor) and 1 μl of Omniscript reverse transcriptase (Qiagen, Omniscript Reverse Transcription, http://www.qiagen.com/). The reaction mixture was incubated at 37°C, for 60 min. Complementary DNA from the RT reaction, 10 × diluted, was used for qPCR. Forward and reverse primers (0.3 μm) were added to 25 μl of PCR SYBR reaction buffer, 2 μl of cDNA and Milli-Q sterile water up to 50 μl of the total reaction volume (Qiagen, QuantiTectTM SYBR Green PCR). Quantitative PCR was carried out using the Smart Cycler II (Cepheid, http://www.cepheid.com/) sequence detector. The PCR cycling conditions comprised an initial HotStarTaq polymerase activation step at 95°C for 15 min, followed by 45 cycles at 95°C for 15 sec and 59°C for 30 sec, and 72°C for 30 sec. A melting curve was performed at the end of the PCR reaction to confirm the product purity. Differences in cycle numbers during the linear amplification phase between samples containing cDNA from infected and uninfected plants were used to determine the differential gene expression. The expression detected from the tomato actin gene was used as an internal standard using the following primers: Primers for actin were: 5′-AACTGGGATGATATGGAGAAGA-3′ and 5′-TCTCAACATAATCTGGGTCAT-3′. Primers for PR1 were: 5′-CCGTGCAATTGTGGGTGTC-3′ and 5′-GAGTTGCGCCAGACTACTTGAGT-3′. Primers for Lox-D were: 5′-GGCTTGCTTTACTCCTGGTC-3′ and 5′-AAATCAAAGCGCCAGTTCTT-3′. In all cases duplicate assays were performed by using cDNA samples derived from two sets of independently grown plants for each experiment.
The growth rate of B. cinerea was analyzed by one-way anova using Statgraphycs Plus software for Windows, V.5 (Statistical Graphycs Corp.; http://www.statgraphics.net). Means were separated using Fisher’s least significant difference (LSD) at 95%. Each experiment was repeated twice with at least four replicates per experiment.
This work was supported by grants from the National R&D Plan (AGL2003-08481), from Plan 2004 of Universitat Jaume I for research promotion (P1 1B2004-35) of the Valencian Regional Ministry of Business, University and Science GRUPOS04/029 and the National R&D Plan (AGL2006-12711-C02-01 and AGL2006-12711-C02-02) MOL is the recipient of a long-term pre-doctoral fellowship from grant AGL2003-08481 and IF is the recipient of a long-term pre-doctoral fellowship from the Valencian Regional Ministry of Business, University and Science. The authors are grateful to María Foo, Victoria Pastor, Pilar Troncho and Sergio Alamar for technical assistance. We thank Dr Ann Powell for critical reading of the manuscript.
- 2004) The salicylic acid-dependent defence pathway is effective against different pathogens in tomato and tobacco. Plant Pathol. 53, 65–72. , , and (
- 2002) Abscisic acid determines basal susceptibility of tomato to Botrytis cinerea and suppresses salicylic acid-dependent signaling mechanisms. Plant Physiol. 128, 491–501. , and (
- 1998) Fungal and plant gene expression during synchronized infection of tomato leaves by Botrytis cinerea. Eur. J. Plant Pathol. 104, 207–220. , , and (
- 2007) A permeable cuticle in Arabidopsis leads to a strong resistance to Botrytis cinerea. EMBO J. 26, 2158–2168. , , , , , , and (
- 2005) Signal crosstalk and induced resistance: straddling the line between cost and benefit. Annu. Rev. Phytopathol. 43, 545–580. (
- 1976) A rapid and sensitive method for quantification of microgram quantities of protein utilizing the principle of protein binding. Anal. Biochem. 72, 248–255. (
- 2001) Cell wall metabolism in fruit softening and quality and its manipulation in transgenic plants. Plant Mol. Biol. 47, 311–340. and (
- 1997) A membrane-anchored E-type endo-1,4-beta-glucanase is localized on Golgi and plasma membranes of higher plants. Proc. Natl Acad. Sci. USA, 94, 4794–4799. , , and (
- 1999) Antisense suppression of tomato endo-1,4-beta_glucanase Cel2 mRNA accumulation increases the force required to break fruit abscission zones but does not affect fruit softening. Plant Mol. Biol. 40, 615–622. , and (
- 1996) Pedicel breakstrength and cellulase gene expression during tomato flower abscission. Plant Physiol. 111, 813–820. and (
- 1997) Auxin regulation and spatial localization of an endo-1,4-beta-d-glucanase and a xyloglucan endotransglycosylase in expanding tomato hypocotyls. Plant J. 12, 417–426. , and (
- 2005) Pseudomonas syringae pv. tomato type III effectors AvrPto and AvrPtoB promote ethylene-dependent cell death in tomato. Plant J. 44, 139–154. and (
- 1998) Use of transgenic plants to study the role of ethylene and polygalacturonase during infection of tomato fruit by Colletotrichum gloeosporioides. Plant Pathol. 47, 308–316. , , , , and (
- 2005) Growth of the plant cell wall. Nature, 6, 850–861. (
- 1957) Callose substance in plant cells. Am. J. Bot. 44, 478–488. (
- 1964) Callose formation and subsequent disappearance – studies in ultrasound stimulation. Plant Physiol. 39, 843. and (
- 1989) Transcriptional analysis of polygalacturonase and other ripening associated genes in Rutgers, rin, nor, and Nr tomato fruit. Plant Physiol. 90, 1372–1377. , , and (
- 2002) The role of ethylene and wound signaling in resistance of tomato to Botrytis cinerea. Plant Physiol. 129, 1341–1351. , and (
- 1995) Oligogalacturonides and chitosan activate plant defensive genes through the octadecanoid pathway. Proc. Natl Acad. Sci. USA, 92, 4095–4098. , , and (
- 2001) Abnormal callose response phenotype and hypersusceptibility to Peronospora parasitica in defense-compromised Arabidopsis nim1-1 and salicylate hydroxylase-expressing plants. Mol. Plant Microbe Interact. 14, 439–450. and (
- 2004) Systemic acquired resistance. Annu. Rev. Phytopathol. 42, 185–209. and (
- 2005) Abscisic acid and callose: team players in defence against pathogens? J. Phytopathol. 153, 377–383. , , and (
- 2005) Contrasting mechanisms of defense against biotrophic and necrotrophic pathogens. Annu. Rev. Phytopathol. 43, 205–227. (
- 1996) Differential expression of two endo-1,4-beta-glucanase genes in pericarp and locules of wild-type and mutant tomato fruit. Plant Physiol. 111, 1313–1319. , and (
- 1997) Immunodetection and characterization of tomato endo-[beta]-1,4-glucanase Cel1 protein in flower abscission zones. Plant Physiol. 114, 1541–1546. , and (
- 1986) Characterization of the hydrolytic activity of avocado cellulase. Plant Cell Physiol. 27, 541–552. and (
- 1984) Pea xyloglucan and cellulose. II. Hydrolysis by pea endo-β-1,4-glucanases. Plant Physiol. 75, 605–610. , and (
- 2007) Impairment of cellulose synthases required for Arabidopsis secondary cell wall formation enhances disease resistance. Plant Cell, 19, 890–903. , , et al. (
- 2000) Movement of plant viruses is delayed in a beta-1,3-glucanase-deficient mutant showing a reduced plasmodesmatal size exclusion limit and enhanced callose deposition. Plant J. 21, 157–166. and (
- 1984) Callose deposition during gravitropism of Zea mays and Pisum sativum and its inhibition by 2-deoxy-D-glucose. Planta, 161, 20–26. and (
- 2002) The Arabidopsis thaliana–Pseudomonas syringae interaction. The Arabidopsis Book. DOI: 10.1199-TAB.0039 , and (
- 2000) Linear beta-1,3-glucans are elicitors of defense responses in tobacco. Plant Physiol. 124, 1027–1038. , , , , , and (
- 1992) Post harvest evaluation of transgenic tomatoes with reduced levels of polygalacturonase: processing, firmness and disease resistance. Postharvest Biol. Technol. 1, 241–255. , , , , and (
- 2002) Effects of inhibiting phenolic biosynthesis on penetration resistance of barley isolines containing seven powdery mildew resistance genes or alleles. Physiol. Mol. Plant Pathol. 61, 41–51. , and (
- 1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227, 680–685. (
- 1994) Two divergent endo-beta-1,4-glucanase genes exhibit overlapping expression in ripening fruit and abscising flowers. Plant Cell, 6, 1485–1493. , and (
- 1998) Transgenic analysis of tomato endo-beta-1,4-glucanase gene function. Role of cel1 in floral abscission. Plant J. 13, 303–310. , , , and (
- 2004) Phylogenetic analysis of the plant endo-β-1,4-glucanase gene family. J. Mol. Evol. 58, 506–515. , and (
- 2007) Overexpression of pectin methylesterase inhibitors in arabidopsis restricts fungal infection by Botrytis cinerea. Plant Physiol. 143, 1871–1880. , , , , , , and (
- 2006) Significance of inducible defense-related proteins in infected plants. Annu. Rev. Phytopathol. 44, 7. , and (
- 1972) Phloem translocation and heat-induced callose formation in field-grown gossypium-hirsutum-l. Plant Physiol. 50, 366. (
- 1995) Nature and regulation of pistil-expressed genes in tomato. Plant Mol. Biol. 28, 691–711. and (
- 2000) Evidence that endo-1,4-beta-glucanases act on cellulose in suspension-cultured poplar cells. Plant J. 24, 147–158. , , , , , and (
- 2003) Feedback from the wall. Curr. Opin. Plant Biol. 6, 611–616. and (
- 2004) Characterization of tomato endo-beta-1,4-glucanase Cel1 protein in fruit during ripening and after fungal infection. Planta, 220, 80–86. , , , and (
- 1999) Cooperative disassembly of the cellulose-xyloglucan network of plant cell walls: parallels between cell expansion and fruit ripening. Trends Plant Sci. 4, 176–183. and (
- 1996) Systemic acquired resistance. Plant Cell, 8, 1809–1819. , , , , and (
- 2007) Mutations in LACS2, a long chain acyl-CoA synthetase, enhance susceptibility to avirulent Pseudomonas syringae, but confer resistance to Botrytis cinerea in Arabidopsis. Plant Physiol. DOI: 10.1104/pp.106.094318. , and (
- 2004) beta-amino-butyric acid-induced resistance against necrotrophic pathogens is based on ABA-dependent priming for callose. Plant J. 38, 119–130. and (
- 2002) PMR6, a pectate lyase-like gene required for powdery mildew susceptibility in Arabidopsis. Plant Cell, 14, 2095–2106. , , and (
- 2004) The role of plant cell wall polysaccharide composition in disease resistance. Trends Plant Sci. 9, 204–209. , and (
- 2003) In spite of induced multiple defense responses, tomato plants infected with cucumber mosaic virus and D satellite RNA succumb to systemic necrosis. Mol. Plant Microbe Interact. 16, 467–476. , and (
- 1995) The tomato Never-ripe locus regulates ethylene-inducible gene expression and is linked to a homolog of the Arabidopsis ETRZ gene. Plant Physiol. 107, 1343–1353. , , , , and (
- 2000) Potentiation of pathogen-specific defense mechanisms in Arabidopsis by beta-aminobutyric acid. Proc. Natl Acad. Sci. USA, 97, 12920–12925. , , and (
- 2001) Beta-aminobutyric acid-induced protection of Arabidopsis against the necrotrophic fungus Botrytis cinerea. Plant Physiol. 126, 517–523. , and (