Using dynamic diversity to achieve durable disease resistance in agricultural ecosystems

Abstract

Plant Pathology Group, Institute of Integrative Biology, ETH Zurich, 8092 Zurich, SwitzerlandE-mail: bruce.mcdonald@usys.ethz.chGenetically resistant cultivars form the foundation of disease management for most crops. Populations of resistant cultivars grown at the farm and landscape scales keep many important fungal, bacterial and viral diseases in check, sometimes for many years. But eventually, most resistant cultivars fail and an epidemic ensues, often causing significant losses in crop yield and quality. In this letter, disease resistance is considered as a population phenomenon, exhibited mainly at the scale of fields, farms and landscapes and encompassing millions or billions of plants, rather than at the level of individual plants or specific plant tissues, even though cultivar resistance is often measured based on symptoms of individual plants or tissues. But it should be noted that the failure of resistance at the population or landscape level generally begins with a failure of resistance at the individual level.Disease resistance fails because pathogens evolve. Evolution requires genetic diversity. Genetic diversity is affected by mutation, population size, recombination, gene flow, and selection, the same factors that affect an organism’s population genetics. Thus the keys to developing strategies for breeding durable disease resistance lie in understanding pathogen population genetics. The root of the problem of pathogen evolution is the lack of diversity in agricultural ecosystems (agroecosystems). Since the invention of agroecosystems ~12,000 years ago, crop genetic diversity has declined steadily in agroecosystems globally to facilitate gradual improvements in agricultural production systems, including innovations such as tillage, fertilization, controlled irrigation, and mechanization. The decline in crop genetic diversity at the field scale accelerated rapidly during the last 100 years to increase the efficiency of food production and feed the burgeoning human population. The large-scale mechanization of planting and harvesting operations and the green revolutions that swept through agriculture led to replacement of locally-adapted but genetically diverse land races by genetically uniform but broadly-adapted, high-yielding, dwarf cultivars, further depleting genetic diversity and increasing the environmental homogeneity present in agroecosystems worldwide. The highly mechanized modern industrial agroecosystems now found around the world are extremely productive food factories, but also are highly effective incubators of pathogen evolution (as described in Stukenbrock & McDonald, 2008). As a result of the increased planting density and genetic uniformity of host populations in agroecosystems compared to natural ecosystems, pathogen transmission became easier, enabling an increase in pathogen virulence (Read, 1994). The same factors increased pathogen population sizes, which led to more genetic diversity for selection to act upon by increasing the total number of mutations available at the field scale while simultaneously lowering the effects of genetic drift. Due to these changes in agroecosystems over time, pathogen evolutionary potential likely increased as agricultural pathogens (including fungi, bacteria and viruses) became domesticated and adapted to the agroecosystem environment. Other factors unique to agroecosystems are also likely to contribute to more rapid pathogen evolution. The higher host density found in agroecosystems compared to natural ecosystems allows agricultural pathogens to exist at a higher density. This higher pathogen density increases the likelihood of multi-infections in the same plant by different genotypes of the same pathogen (e.g. Linde et al. 2002; Keller et al. 1997; McDonald et al. 1999). Multi-infections have long been thought to favor the development of higher virulence (defined here as the amount of damage done to the plant, but often referred to as aggressiveness or quantitative virulence in plant pathology) as a side effect of competition among strains for host resources (Van Baalen & Sabelis, 1995). The higher host and pathogen density found in agroecosystems also increases the likelihood of co-infection by different pathogen species, which also is thought to select for higher virulence as a result of competition among pathogen species for the same host resources. In addition, co-infection by different pathogen species increases the likelihood of horizontal transfer of genes among pathogens infecting the same host, as reported recently for several wheat pathogens (e.g. Friesen et al. 2006; Gardiner et al. 2012; McDonald et al. 2013).Modern agroecosystems are very well structured to produce high quality produce that can be eaten directly (e.g. rice or apples) or processed into higher value goods