Following clues from the molecular biology of plant-microbe interactions, many genes involved in disease resistance and the defense response have been cloned from a variety of plants (Dangl and Jones 2001). A number of these genes have been tested for their ability to control fungal pathogens in transgenic plants grown in the laboratory and to a more limited extent in the field. In addition, natural or synthetic antimicrobial peptides and genes for resistance to pathogen-derived toxins have been introduced into plants. This work is thoroughly reviewed by Bent and Yu (1999); Rommens and Kishore (2000), and Melchers and Stuiver (2000). Only more recent studies providing significant new findings or extensions of this work will be considered here.
Current efforts have continued to focus on introduction of disease resistance genes, natural and synthetic antimicrobial peptides as well as selected enzymes such as chitinase and b-glucanase into several crop plants and the laboratory and field evaluation of these plants for disease resistance. Little has been reported in the literature on the field performance of these transgenic plants, although about 5% of all permits for field testing of transgenic plants in the United States over the last decade have been for transgenic plants having fungal resistance (Information Systems for Biotechnology 2002). Moreover, most permits have been issued to companies and this proprietary work is not yet in the public domain. No transgenic plants having fungal resistance have been approved for use as food and/or feed (AgBios 2002). Despite this seemingly limited progress, some promising achievements have been made in the last few years and many major resistance genes associated with fungal disease resistance have been cloned or at least tagged with molecular markers (Brueggeman et al. 2002; Chauhan and Leong 2002; Gebhardt and Valkonen 2001). Thus, we can expect many new developments with regard to this field in the next decade. Furthermore, despite public concern about transgenic crops, the global adoption of transgenic crops continues to increase, particularly in the United States, where 74.8 million acres were planted in transgenic crops including corn, soybean, cotton, and canola in 2000 (Transgenic Crops 2002). This represented about 50% of the total soybean and cotton acreage planted in that year. The International Service for the Acquisition of Agri-Biotech Applications predicts that the world market for genetically engineered plants will be $8 billion in 2005 and $25 billion by 2010 (http://nature.biotech. com) (Figure 3). Plant pathogens cause $30-50 billion dollars of loss annually in crop productivity (Baker et al. 1997) thus justifying this investment in biotechnological approaches to crop protection. The reduction in use of agrochemicals for disease control is another important incentive for this technology. Japanese growers spend more than $600 million a year to control diseases on rice (Bonman 1998). Already, the reduction in insecticide use in China through use of Bt transgenic crops has impacted farmer income and health (Huang et al. 2002; Pray et al. 2002).
2.2.1 Defense Pathways
Despite many projections made in the reviews listed earlier, a flurry of published reports has not followed. In many cases, this can be attributed to the observation being made first in Arabidopsis and not in a crop plant. However, this lag is being addressed with Arabidopsis genes to assess function in crop plants and through the identification of homologs of Arabidopsis genes. For example, the Arabidopsis NPR1 gene (Cao et al. 1998) when overexpressed in rice caused enhanced resistance to the rice pathogen Xanthomonas oryzae pv. oryzae (Chern et al. 2001) and the investigators were able to retrieve, in a two hybrid screen, a bZIP family of interactors showing that a similar pathway of signaling is likely present in rice as Arabidopsis (Zhang et al. 1999). In addition, Yoshioka et al. (2001) have shown that Arabidopsis will respond to the rice fungicide probenazole by induction of PR genes and show enhanced resistance to Pseudomonas syringae pv. tomato DC3000 and P. parasitica Emco5. This response was dependent on a functional NPR1 gene and was compromised in NahG transgenic plants further supporting the connection of this pathway with generalized resistance to pathogens in both Arabidopsis and rice. Six NPR1 homologs are reported in the recently released Nipponbare genome sequence (Goff et al. 2002). It will be interesting to see how overexpression and silencing of these genes affects resistance of rice to key fungal pathogens of rice such as Rhizoctonia solani and M. grisea.
Overexpression of the Arabidopsis ACD2 (accelerated cell death) gene leads to tolerance of susceptible Arabidopsis plants to P. syringae infection by reducing disease symptoms associated with cell death such as ion leakage, while allowing the bacteria to grow to similar levels as in susceptible plants (Mach et al. 2001). Fungal pathogens were not tested.
Broader testing of other genes that have shown wide-spectrum disease resistance to bacterial, fungal, and viral pathogens when overexpressed such as Prf (Oldroyd and Staskawicz 1998) and Pto (Tang et al. 1999) in tomato has not been reported. Nor have further reports been made on constitutively active variants of Pto (Rathjen et al. 1999). Presumably, this approach can be used in other crop plants.
Introduction of the bacterial blight resistance gene Xa21 into elite rice cultivars has lead to the expected resistance phenotype when inoculated with X. oryzae pv. oryzae; however, strains that are virulent on Xa21 were not tested nor were other pathogens (Tu et al. 1998). Performance of these lines was tested under natural field conditions without any apparent loss of yield performance (Tu et al. 2000).
Disease resistance genes from one crop plant have now been successfully used in other crop species. For example, the Bs2 resistance gene from pepper confers resistance to X. campestris pv. vesicatoria in tomato in the laboratory as well as in preliminary field tests (Staskawicz et al. 2002; Tai et al. 1999). Work from my laboratory in conjunction with studies from the laboratories of Mark Farman at the University of Kentucky and Yukio Tosa at Kobe University has suggested that the Pi-CO39 (t) gene (Chauhan et al. 2002)
for resistance in rice to M. grisea carrying the AVR1-CO39 gene (Farman and Leong 1998) will be useful in other grass species such as perennial rye grass as functional copies of the AVRl-CO39 gene are found in the grey leaf spot pathogen (ML Farman personal communication).
Coexpression of the C. fulvum Avr9 and tomato Cf-9 genes in Brassica napus was investigated as a method for inducing broad-spectrum resistance to fungal pathogens (Hennin et al. 2001). Induction of PR1, PR2, and Cxc750 was detected following injection of the Avr9 peptide obtained from intercellular fluids of B. napus transgenic plants expressing Avr9 into B. napus expressing the tomato Cf-9 gene. F1 plants and progeny from a cross of the Avr9 and Cf-9 plants were evaluated for resistance to fungi. Disease development was delayed at the site of infection of L. maculans and Erysiphe polygoni but enhanced at the site of infection of Sclerotinia sclerotiorum. Thus, heterologous expression of AVR-R gene pairs may be a useful strategy for control of fungal disease in a variety of plants. However, the finding that Arabidopsis resistance gene RPM1 requires another plant gene RIN4 in order to accumulate and interact with avrRpm1 or AvrB (Mackey et al. 2002) as well as the inability to show direct interaction of the products of Avr9 with Cf-9 (van der Hoorn et al. 2002; Luderer et al. 2001) suggests that this strategy must be used cautiously. This may explain the inability of van der Hoorn et al. (2002) to see a necrotic response in the nonsolanaceous plant lettuce with this gene combination. More recent reports on the use of the coexpression strategy for plant protection against fungi in solanaceous plants have not been made (Melchers and Stuiver 2000), however the coexpression of Avr9 and Cf-9 under control of nematode inducible promoters in tobacco has been studied (Bertioli et al. 2001). Surprisingly these plants underwent spontaneous necrosis in the absence of the nematode. Evidence for activity of the genes was found both in aerial and root tissue.
The cloning of MLO locus of barley (Shirasu et al. 1999), which confers nonrace-specific resistance to Blumeria graminis f. sp. hordei, has been followed with investigations of its potential use for control of different fungal pathogens. Jarosch et al. (1999) found that in contrast to increased resistance conferred by recessive alleles of MLO to powdery mildew, these barley plants have increased susceptibility to penetration by M. grisea despite showing similar ability to wild type plants to respond to M. grisea elicitor. Likewise Kumar et al. (2001) showed that mlo plants were more susceptible to the necotrophic pathogen Bipolaris sorokini-ana. These reports reveal the complexity of various fungal interactions with the host and the difficulty of using a single strategy to control multiple pathogens.
Recent work on MLO has shown that it is a novel calmodulin-binding protein that is responsive to both abiotic and biotic stresses through down regulating the oxidative burst and cell death response (Kim et al. 2002a,b; Piffanelli et al. 2002). Binding to calmodulin is essential to full function of MLO (Kim et al. 2002a). A rice homolog of MLO that also interacts with calmodulin was isolated (Piffanelli et al. 2002). It will be interesting to see how silencing of MLO in rice plants affects interaction with biotrophic and necotrophic pathogens.
Several new reports have appeared on the use of antifungal proteins such as Ag-AFP from Aspergillus giganteus, chitinase, b-glucanase and Ribozome-Inactivating Proteins (RIP) (Chareonpornwattana et al. 1999; Datta et al. 2002; Oldach et al. 2001), thaumatinlike protein (PR-5) (Datta et al. 1999), and human lysozyme (Takaichi and Oeda 2000) in plants for protection against fungal disease and show different levels of promise for these approaches. Chitinase and AFP appear to increase resistance in wheat, however, the results were not corroborated with levels of these proteins in transgenic plants (Oldach et al. 2001). Other efforts to introduce chitinases in wheat have led to gene silencing (Chareonpornwattana et al. 1999). The recent isolation of cDNA clones for novel acidic chitinases and b-1,3-glucanases from wheat spikes infected by Fusarium graminearum (Li et al. 2001) is exciting as these enzymes may be more effective in control of this pathogen in this tissue. Introduction of infection-related chitinase and rice thaumatinlike protein into rice has led to moderate control of sheath blight caused by R. solani (Datta et al. 2002). Field evaluation of these plants is underway. Finally, studies on carrot transformed with human lysozyme, which can cleave b-1,4 glycosidic bonds of peptidoglycan in bacterial cell walls and chitin in fungal cell walls, suggest that this approach may have promise for control of E. heraclei and A. dauci (Takaichi and Oeda 2000).
Natural and synthetic peptides have been evaluated for control of pre and postharvest damage by fungi. Ali and Reddy (2000) studied four cationic peptides for antimicrobial activity in vitro and in plants. All were shown to have significant activity in the micromolar range against P. infestans and A. solani completely inhibiting growth of the fungi on potato tissues. Alfalfa antifungal peptide defensin from seeds of Medicago sativa was shown by Gao et al. (2000) to have significant activity against Verticillium dahliae in vitro, and transgenic potato plants expressing the peptide showed a reduced area under the disease progress curve compared to vector control plants. Moreover, resistance was correlated with the levels of peptide found in root samples.
Similar results were obtained for transgenic potato plants expressing a N terminus-modified cecropin-melittin cationic peptide chimera (Osusky et al. 2000). The efficacy of the peptide against Phytophthora cactorum and F. solani infection was demonstrated in variety Desiree while not affecting plant growth or tuber morphology or size. Tubers remained resistant for more than one year and the peptide could be detected in this tissue. By contrast transgenic Russet Burbank plants showed significant morphological alterations and resembled lesion mimic plants, produced very small tubers, and showed less resistance to P. cactorum. Trangenic raw tubers were fed to mice without significant growth effects relative to untransformed tubers. Rajasekaran et al. (2001)
have shown that the synthetic antimicrobial peptide DE41 is active at the micromolar range against many important bacterial and fungal plant pathogens. Crude protein extracts from transgenic tobacco plants constitutively expressing D4E1 showed ability to reduce growth of A. flavus and V. dahliae while control plant extracts did not (Cary et al. 2000). Furthermore, the transgenic plants showed increased resistance to Colletotrichum destructivum. The D4El gene has been introduced into cotton and was shown to be present in cottonseed. Reduction of aflatoxin in cottonseed oil is a desired outcome. Dow AgroSciences LLC has licensed the technology and is collaborating with USDA-ARS scientists who developed the technology to further evaluate the efficacy of the transgenic plants (SeedQuest 2002). The effects of this peptide on plant growth or other nontarget organisms have not been reported. Finally, the antimicrobial peptide MS1-99, an analog of magainin 2, a defense peptide secreted from the skin of the African clawed frog, was expressed from the chloroplast genome of tobacco and showed significant ability to control many phytopathogenic bacteria and fungi (DeGray et al. 2001). Trangenic plant homogenates inhibited the fungi A. flavus, F. monilzjbrme and V. dahliae and anthracnose lesions were absent in transgenic plant infected with C. destructivum. Transformation of the chloroplast genome is an innovative approach to control the spread of the transgene as pollen will not carry the transgenic chloroplast. Evidence for pollen transfer of transgenes at the commercial field level is now available (Reiger et al. 2002).
While the use of peptides has shown significant promise, thorough testing of the toxicity of plants producing these peptides will be important. Their potential ability to inhibit microflora that are essential to plant health as well as health of animals, humans, and birds needs careful evaluation. Feeding raw potatoes containing a cecropin-melittin cationic peptide chimera to mice was not a convincing test of the toxicity as the animals do not normally eat this food and lost weight on this diet until it was supplemented with normal feed (Osusky et al. 2000). More realistic tests are needed. The use of cooked potatoes should be tested. Furthermore, only short-term effects were studied. The survival and biological impact of large quantities of these peptides in the environment resulting from crop plant decay also needs to be evaluated. These issues are only beginning to be addressed in a multitrophic context for insect resistant transgenic plants (Groot and Dicke 2002). These issues need to be critically addressed at the time of risk assessment. Public concern has been spurred largely by a lack on confidence in transgene technology because of the lag time in responding to the large-scale effects that agrochemicals are having on human and environmental health despite early warnings by Carson (1962) many decades ago.
In planta studies of the hydroxylation and glycosylation of destructin B, a phytotoxin produced by A. brassicae, to a nontoxic product have shown a correlation between plant resistance with phytoalexin production and the efficiency of these modifications of the toxin in Brassica spp. (Pedras et al. 2001). These data suggest that improved resistance can be engineered in or transferred within Brassica hosts of A. brassicae by enhancing hydroxylation of destruxin B.
The analysis of the biosynthetic pathway of saponins, antimicrobial metabolites of plants, may allow the transfer of these genes to other plants. Mutants defective in the saponin avenacin in oat were studied and shown to define seven loci and to be compromised in their ability to resist fungal attack (Haralampidis et al. 2001). The sad1 gene was shown to encode b-amyrin synthase.
Was this article helpful?