Department of Plant Physiology, University of Kaiserslautern, Kaiserslautern, Germany
Disease resistance in crop plants is a major challenge in plant breeding. Within industrialized agriculture, the need for efficient self-protection of cultivated plants is likely to increase in the future. To achieve this goal, diverse strategies will be necessary and compete. Conventional breeding has made great progress in incorporating natural defense genes, but the limitations of this method are also obvious. The progress in plant molecular biology now allows the generation of transgenic plants, thereby exploiting the mechanisms nature has developed to control and to limit the infection of microbial pathogens on plants. An additional benefit of transgenic plants is the possibility to validate the usefulness of endogenous plant genes in defense strategies. Changes in the expression or composition of a desired gene may still be achieved by conventional breeding assisted by the results from transgenic plants.
The possibility to transform all major crop plants—although sometimes tedious and labor intensive—opens a new opportunity for novel enhanced plant tolerance toward microbial pathogens. Over the last few decades many of the natural plant defense strategies were thoroughly investigated, and as a result numerous mechanisms are known by now. For example, the concept of pathogenesis-related proteins (PR-proteins) has revealed an inducible defense system that is thought to have antimicrobial properties. The constitutive expression of a single or a few genes of this group of defense genes was started in the late 1980s but with limited success. This will be covered in the first section of this chapter. On the other hand, new strategies for enhanced pathogen tolerance try to use the plants' signaling network for defense activation. As a consequence, a whole battery of diverse defense reactions is triggered (variable in different plant species), which seems to be more powerful than the overexpression of single antimicrobial proteins. These approaches acting on complex defense networks will be described in the second part.
I. SINGLE-GENE DEFENSE MECHANISMS A. Pathogenesis-Related Proteins
About 40 years ago Ross (1,2) performed experiments with tobacco Xanthi nc plants, which after viral inoculation exhibit a hypersensitive reaction in which a limited number of cells die and form a lesion. Plants that were inoculated with the tobacco mosaic virus showed much smaller lesions after a challenge infection 7 days later than newly inoculated leaves. This phenomenon is now well known as systemic acquired resistance (SAR). Searching for the mechanisms underlying SAR, PR-proteins were discovered (3). Subsequently some of them were biochemically identified as hydrolases of fungal cell walls, namely /3-1,3-glucanase (4,5) and chitinase (6). However, the biochemical properties and enzymatic function of other PR-proteins such as the well-characterized PR-1 remain puzzling (7).
Transgenic tobacco plants overexpressing a chitinase gene from tobacco showed enhanced tolerance against fungal infection by Cercospora nicotianae (8) indicating that high levels of plant hydrolyzing enzymes are a suitable strategy to increase pathogen tolerance. By using a combination of a /3-1,3-glucanase and a chitinase gene, Broglie et al. (9) demonstrated the synergistic effect of both enzymes, yielding a higher degree of resistance compared with the expression of each gene alone. A similar conclusion was drawn from a study in tobacco, in which a gene for a basic chitinase from rice and a gene for an acidic glucanase from alfalfa were coexpressed after appropriate crossing of individual transgenic plants (10). The different gene combinations were compared, as was the level of expression of these hy-drolytic enzymes in homozygous versus heterozygous plants. Based on the reduction of lesion size after C. nicotianae infection, it was shown that the combination of glucanase and chitinase expression even at a moderate level is more beneficial than the expression of either gene alone at a much higher level (10). The concept of the constitutive expression of hydrolytic enzymes was later extended from tobacco to crop plants. The expression of chitinase in oilseed rape (Brassica napus) to obtain enhanced tolerance was tested in field trials after inoculation with three different fungal pathogens (11). Although the overall protection of the transgenic plants in the field trials seems to be smaller than previously shown in experiments in a greenhouse, the enhanced tolerance in chitinase-expressing rape was effective against several fungi under natural field conditions. Many other plant species were transformed with plant glucanases or chitinases, and most research groups were able to find at least one fungus that is sensitive to these hydrolytic enzymes (12-14).
The reports on transgenic plants overexpressiong hydrolytic enzymes imply that this strategy might already be sufficient to combat most fungal pathogens. However, this is obviously not the case. Quite often fungi were chosen for pathogenicity assays that are known to be sensitive to chitinases and/or glucanases, Rhizoctonia solani and Trichoderma viride being examples. In contrast, well-recognized plant pathogens are often not significantly inhibited in their pathogenicity (15). Thus, plant hydrolases are useful to limit the spread of some fungal pathogens and are likely to increase the basal tolerance of plants against fungal infections. Whether the increased tolerance of transgenic plants with elevated levels of hydrolytic enzymes is the result of the direct inhibition of the fungal (tip) growth needs to be carefully analyzed in future work. Enzymatic cleavage of fungal cell walls liberating chitin or glucan oligomers will also activate diverse plant defense responses as these carbohydrate oligomers are potent elicitors in almost every plant (16,17). The transcriptional activation of genes involved in lignin precursor formation or antimicrobial phytoalexins by elicitors is well known (18,19). The genes for PR-proteins of unknown biochemical function such as PR1 or PR5 were also constitutively expressed in tobacco plants. The PRla-overexpressing plants have a higher degree of resistance against a limited number of fungal pathogens from oomycetes (20). However, no enhanced tolerance against other pathogenic fungi of tobacco or tobacco mosaic virus was achieved. It was later shown that PR 1-proteins from tobacco and tomato inhibit the spore germination of Phytophthora infestans and reduce the lesion size of diseased tomato leafs (21). Notably, the PR la isoform, used in the transgenic lines (20), was very inefficient and showed only 10% of the biological activity of the most potent antifungal isoform
The overexpression of a rice thaumatin-like gene of the PR5 family in rice plants gave enhanced tolerance against Rhizoctonia solani, the agent causing sheath blight disease. The infected leaf area was reduced to one fifth in the best lines, indicating that the thaumatin-like gene is highly effective against R. solani infection (22). Expression of a similar osmotin-like gene in potato also enhanced tolerance against the late blight disease caused by Phytophthora infestans (23).
One interesting point about PR-proteins in cereals is worth mentioning. Wheat, for instance, expresses hydrolytic enzymes (/3-1,3-glucance, chitin-ase) after pathogen infection. In contrast to this situation in Arabidopsis and tobacco, the status of a systemic acquired resistance in wheat is not correlated with the constitutive expression of these PR-genes (24). This raises the question of whether hydrolytic enzymes are an important part of the plant defense system at all. Alternatively, other as yet unidentified genes (or mechanisms) are important players in plant tolerance to pathogens. Support for the latter view comes from experiments with Arabidopsis DNA microarrays in which many genes are transcriptionally induced in resistant plants (25). The fact that 413 genes (out of —7000 genes representing 25% of the genome) were reported to show a consistently higher expression level in SAR or plant resistance indicates that there will be more than a thousand genes that are significantly induced in local resistance or SAR. This incredibly high number draws a far more complex picture of plant resistance than previously thought. As well as being frustrating for researchers trying to dissect single-gene function in SAR, the high number of SAR-related genes is a huge challenge and opportunity for the plant molecular biologist to create novel transgenic lines with enhanced tolerance against a variety of microbial pathogens.
Plant and animals have developed an efficient mechanism to combat pathogens by using small antimicrobial peptides collectively termed defensins. These peptides are now divided into several families on the basis of sequence homology and structural properties. They are usually relatively small (<60 amino acids) and in the case of plant defensins contain several cysteine residues that form one or more stable disulfide bridges. Defense peptides exhibit a broad spectrum of antimicrobial activity against bacteria, fungi, and even enveloped viruses. In animal cells even parasites and tumor cells are inhibited (26). Most of the known defensin genes are from insects showing mainly antibacterial activity (27,28). The large diversity of these peptides (more than 500 varieties are known) provides a huge reservoir of genes that can be expressed in plants to enhance tolerance against pathogens, an approach that has just started to be exploited by plant molecular biologists.
A plant defensin from radish (RsAFLP2) was expressed at high levels in tobacco plants. Subsequent infection with the fungal pathogen Alternaria longipes revealed efficient protection of the transgenic plants resulting in more than 80% reduction of lesion sizes (29).
Lipid transfer proteins were initially described as shuttle proteins, involved in the transfer of lipids between organelles (30). Later, the antimicrobial activity of these peptides was discovered (31,32). Transgenic Arabidopsis plants overexpressing the lipid transfer protein LTP2 from barley showed a strong reduction in disease symptoms after infection with Pseudomonas syringae pv. tomato (33).
Thionins are often found in quite high amounts in the endosperm of cereals and other plants. They are toxic to plant pathogenic fungi but only at relatively high concentrations. Bohlmann et al. (34) showed inhibition of the barley pathogen Drechslera teres at concentrations of 500 /xM. Nevertheless, transgenic Arabidopsis plants expressing high thionin levels were shown to be more resistant to Fusarium oxysporum (35) and Plasmodi-ophora brassicae (36).
The overexpression of plant defense peptide genes to achieve higher tolerance against a broad range of pathogens was only partly successful. A few pathogens (mainly bacteria) were restricted in their growth, but many other pathogens were not. This problem was overcome by the expression of chimeric defense peptides in potato plants. The synthetic peptides consist of two domains derived from cecropin (from the giant silk moth Hyalaphora cecropia) and melittin (the major component of bee venom) (37). The N-terminus of the chimeric peptide had to be modified in order to be tolerated as nontoxic by the plant cells. The transgenic potato plants were almost totally resistant against Erwinia carotovora even over a very long period of time (e.g., 6 months of tuber storage). This high degree of resistance was also observed after infection of potato plants with different fungi including Phytophthora cactorum and Fusarium solani (37). As mentioned before, some defense peptides are also toxic to animals, implying the need for great care in the use of these defense molecules in edible plants. The transgenic potatoes containing the chimeric cecropin-melittin peptide were fed to mice for several weeks without any notable change in animal behavior or body weight. The large number of known potent defense peptides from insects combined with molecular biology tools will make it possible to exploit these natural defense mechanisms on a broad basis.
Ribosome-inactivating proteins (RIPs) are widely found in plants. They exhibit a specific RNA-iV-glycosidase activity that selectively cleaves off an adenine residue from a conserved site of the 28S rRNA. This prevents binding of the elongation factor 2 and consequently leads to an arrest in protein biosynthesis (38). RIPs do not inactivate the ribosomes of their own species but inactivate those of distantly related species. Expression of the barley seed RIP in tobacco under the control of a wound-inducible promoter resulted in enhanced tolerance against fungal infections with Rhizoctonia so-lani (39). However, attempts to express the barley RIP in wheat plants were unsuccessful, indicating that the constitutive expression of this protein is toxic for wheat to allow regeneration of transgenic lines (14). RIPs are primarily antiviral proteins, and overexpression of RIPs often leads to resistance of the transgenic plant against a broad spectrum of plant viruses. However, an additional antifungal activity of RIP was observed when overex-pressing pokeweed antiviral protein and mutant forms, which exhibit no /V-glycosidase activity typical of the RIP function (40). The advantage of the mutant RIPs is their low toxicity compared with the original protein.
D. Plants with Elevated Levels of Antimicrobial Secondary Compounds
The efficient protection of many wild-type plants against microbial pathogens is thought to be mediated at least in part by toxic secondary metabolites from plants (41,42). Often these compounds require very complex biosynthesis, and many of the genes involved in their formation are not known or characterized. These limitations usually make the formation of secondary metabolites with complex biosynthesis in transgenic plants very difficult. Therefore, today's strategies are based on the expression of a single gene (or a few genes) to equip a plant with a novel secondary metabolite.
A successful example is the expression of a stilbene synthase gene from grapevine (Vitis vinifera) in tobacco and crop plants under the control of its native pathogen-inducible promoter. The enzyme requires only one p-coumaroyl-coenzyme A (CoA) (a lignin precursor) and three malonyl-CoA to form one molecule of the stilbene resveratrol, which has antimicrobial properties in plants. The novel phytoalexin accumulated after Botrytis cinerea infection to low millimolar levels within a few days, resulting in a reduction of diseased leaf area of roughly two thirds (43). The same gene was later expressed in tomato (44), rice (45), and other crop plants. Again, a strong increase in basal pathogen tolerance was obtained. One drawback, however, is the observation that tomato plants producing resveratrol showed an increase in resistance against Phytophthora infestans (—50% reduction in diseased leaf area) but not against Botrytis cinerea (44), which was efficiently restricted on resveratrol-producing tobacco plants. In retrospect, it is amazing to see that the grapevine stilbene synthase promoter is pathogen inducible in so many plants and that this inducibility is essential as the constitutive expression of the stilbene synthase gene results in detrimental effects such as male sterility in tobacco (46).
Was this article helpful?