1Soil Fertility and Plant Nutrition Group, PO Box 8003, 6700 EC Wageningen The Netherlands; 2Ecological Phytopathology Group, Agricultural University, PO Box 8026, 6700 EE Wageningen, The Netherlands

Background and objectives

Evolutionary theories on plant defence strategies relate plant growth rate to defence. The resource availability hypothesis predicts that low nutrient or light availability leads to slow inherent growth rate and high investment in defence-related secondary plant compounds. The link between plant growth rate and defence is explained by the premise that defence is costly in terms of fitness. Consequently, better defended plants should grow slower in the absence of herbivores and pathogens. The carbon/nutrient balance hypothesis, predicts that plants from infertile soils have higher carbon/nitrogen (C/N) ratios and higher concentrations of secondary plant compounds consisting of C, H and O only. These plants would be better protected. Both theories have exclusively been tested for plant-herbivore interactions. We evaluated their applicability to plant/pathogen interactions.

Materials and methods

Through literature review we tried to find evidence for the premise that resistance is costly. To test the RA hypothesis, growth rate and susceptibility to Fusarium oxysporum of radish cultivars were analysed. Growth rates of mutants of Arabidopsis thaliana known to differ in pathogen susceptibility, were also determined. The CNB hypothesis was tested by growing various plant species at different nitrogen uptake rates. Their resistance towards several, pathogens was tasted.

Results and conclusions
Evaluation of the costs of resistance is hampered by the lack of proof for secondary plant compounds to be involved in resistance [1]. The best-described compounds are receptor molecules in gene-for-gene systems. Based on mechanistic considerations we expect this highly specific resistance not to be costly. Growth rate analyses of Arabidopsis mutants confirmed this. Other resistance compounds, such as phytoalexins and pathogenesis-related proteins, are synthesized only after attack and the type of resistance provoked should not be costly either. Phenolics are the best-known constitutive compounds involved in nonspecific resistance. Their effect is concentration-dependent and resistance based on phenolics should be costly. Experiments confirmed that radish cultivars with higher phenolic concentrations grow slower and are more resistant [2]. We conclude that the RA hypothesis sensu sticto is not applicable to plant-pathogen interactions, although in some type of plant-pathogen interactions slow inherent growth correlates with high resistance. The test of the CNB hypothesis demonstrated that a plant's, resistance can strongly depend on its C/N ratio. Tomato plants with a high C/N ratio are more susceptible to Botrytis cinerea (unpubished data). Probably, levels of soluble carbohydrates and not levels of resistance-related compounds explain the positive correlation between susceptibility and C/N ratio. Cauliflower and pea plants with a high C/N ratio are more resistant to downy mildew (Peronospora sp.). Thus, resource availability can strongly affect pathogen resistance phenotypically through the plant's chemical composition.

1. Hoffland E, Jeger MJ, van Beusichem ML, 1998. Is plant growth rate related to disease resistance? In: Lambers H, Poorter H, van Vuuren MMI, eds. Inherent variation in Plant Growth. Physiological Mechanisms and Ecological Consequences. Leiden: Backhuys Publishers, in press.
2. Hoffland E, Niemenn GJ, van Pelt JA, Pureven JBM, Eljkel GB, Boon JJ, Lambers H, 1996. Plant, Cell and Environment 19, 1281-1290.