A central player in many but not all forms of programmed cell death in plants is salicylic acid (76). This was first demonstrated in transgenic tobacco and Arabidopsis plants expressing a salicylate hydroxylase gene from Pseudomonas putida called nahG (77). These plants are still capable of synthesizing SA, but as soon as SA starts to accumulate it is converted into catechol by the enzyme salicylate hydroxylase. Arabidopsis plants of the Col-O eco-type, which as wild-type plants are genetically resistant to Peronospora parasitica infection, were totally colonized by the fungus in nahG-expvess-ing plants (76). This conversion of an incompatible to a compatible interaction could be reverted by spraying with high concentrations of salicylic acid or the synthetic analogue INA (2,6-dichloro-isonicotinic acid), indicating that SA is a key control molecule in plant resistance. Some of the dwarf lesion mimic mutants of Arabidopsis mentioned before could also be re verted to wild-type-like plants when crossed with nahG plants (78). Thus, the question arises of how SA controls programmed cell death in plants.
The discovery tour begins with a binding protein for SA, characterized from tobacco, that was shown to be a catalase (79), pointing to increased levels of H202 as the driving force of programmed cell death. This view was later extended by the finding that the other major H202-scavenging enzyme, ascorbate peroxidase, can also be inhibited in tobacco by SA (80). This model is attractive at first glance as H202 is an important signaling molecule in plant defense, as outlined earlier (53). However, many other groups have tested the phenomenon in various plant species, and none of their studies has shown a significant inhibition of catalase or ascorbate peroxidase at physiologically relevant concentrations (<100 ¿lM). For example, in soybean SA (100 /xM) neither inhibits both enzymes in vitro nor interferes with the capacity of soybean cell cultures to metabolize H202, which reflects the in vivo situation (81).
In a model system for programmed cell death in plants, we investigate the HR in soybean cell cultures (cv. Williams 82) triggered by the avrA protein of the bacterial pathogen Pseudomonas syringae pv. glycinea. The HR occurs according to the gene-for-gene hypothesis requiring the Rps2 resistance gene in soybean and the avrA avirulence gene in Pseudomonas. The advantage of the cell culture system is its flexibility and the possibility to manipulate the system easily using inhibitors or activators. The HR requires SA (—50 ¿¿M), which can be added to the cell culture medium to complete the cell death program. This phenotype in soybean resembles the studies with nahG Arabidopsis plants, in which the HR is also dependent on SA. The addition of SA to soybean cell cultures strongly accelerates the HR, finally leading to faster cell death (82). By using the vital stain Evans Blue, we showed loss of membrane control as the final sign of cell death in the HR after about 8 hours in the SA-accelerated HR compared with about 14 hours in control inoculated cells without SA (similar to the time range of the HR in plants).
Using this acceleration for a screenable test system, we identified several chemicals that can substitute for SA (83). To our surprise, the chemicals were diverse in structure and thus made it initially difficult to come up with a model to explain their biochemical function. We have collectively termed these chemicals FASs (functional analogues of SA). A rigorous check of the literature revealed an emerging function of FAS compounds in animal cells. Almost all of them are ligands of a transcription factor system in humans, the peroxisome proliferator-activated receptors (PPARs). The PPARs are a subfamily within the superfamily of nuclear hormone receptors, which are posttranslationally activated by binding their specific ligands (84). One therapeutic aspect of PPARs in humans is their relevance in type II diabetes, which can be treated by the uptake of synthetic PPAR ligands (such as troglitazone) on a daily basis (85). This drug can also be used in soybean to complement and accelerate the HR, triggered by Pseudomonas bacteria, and by this criterion is an FAS compound. The outlined model predicts that genes in plants are transcriptionally induced by SA or FAS compounds. We have described a novel putative lectin-encoding gene from soybean that is induced by both groups of compounds as predicted (83). Using differential display and subtractive suppression hybridization, we have identified a larger set of genes, which are transcriptionally induced by salicylic acid and FAS compounds (C Anstätt, N Ausländer, A Ludwig, G Schwerdtfeger, M Hansen, R Tenhaken, unpublished results). This set probably includes novel genes, which are required for the execution of programmed cell death in a salicylic acid-dependent manner.
Chemicals such as INA or the benzothiadiazole Bion® are thought to mimic SA and are thus able to activate the signaling process leading to SAR (86). This raises the question of whether FAS chemicals are similar to SAR inducers in their mode of action. Arabidopsis and tobacco treated with SA or Bion show a strong induction of the PR-1 messenger RNA (mRNA) (Fig. 1). In contrast, the FAS chemical flufenamate was unable to induce the same response in tobacco, pointing to different signal transduction pathways. In the model plant tobacco, several SA-inducible genes were described besides the classical PR genes, for instance, in a study by Horvath and Chua (87,88). Whether theses genes are novel PR genes in the sense that they are induced in plant-pathogen interactions remains to be analyzed. We have chosen the C14-lb (88) gene as a probe to test whether FAS compounds will induce genes in tobacco as well as in soybean. The sensitive induction of the tobacco C14-lb gene by flufenamate, which occurs at a much lower concentration compared with SA, is an example (Fig. 2). As also shown in Fig. 2, another tobacco gene, G8-1 (88), is strongly induced by salicylic acid but almost not induced by the FAS compound flufenamate. These observations underline the different signal transduction pathways and mode of actions in which salicylic acid is involved in plants. The data can be summarized in a new model in which SA controls plant resistance by at least two independent processes. In the first place, high concentrations of SA are needed to execute the plant cell death program, and the predicted mode of action is the transcriptional activation of genes (which mostly remains to be identified in future studies). This function of SA can be mimicked by FAS compounds as shown for soybean. The second major function of SA is the participation in the establishment of the plant immunity (SAR). This function of SA can be potently mimicked by INA or Bion.
Our current goal is to identify an SA/FAS-regulated transcription factor that is involved in programmed cell death in plants. By functional analysis
Was this article helpful?