Screening for costs of disease resistance caused by stomatal dysfunction

Abstract

Wheat varieties combining the highest yields and good resistance against septoria and rusts have proven elusive. There is significant evidence that some disease resistance genes impose a yield penalty. Hence, breeding for disease resistance creates ‘yield drag’ which slows the rate of yield improvement. Breeding for disease resistance is a high priority for plant breeders, and it would therefore be beneficial to select effective resistance genes during varietal development that do not exhibit a yield cost, to maximise yield potential.

The aim of this project was to identify resistance genes or QTL which do, or do not, exhibit a yield penalty and develop methods to minimise ‘yield drag’ associated with breeding for disease resistance. Specifically four deliverables were addressed. (1) Quantify yield penalties associated with resistance genes/QTL effective against Zymoseptoria tritici (septoria tritici blotch), Puccinia striiformis (yellow rust) and Puccinia triticina (brown rust); (2) Quantify yield penalties associated with ‘defeated’ resistance genes/QTL; (3) Identify/optimise methods to screen future resistance genes/QTL for yield penalties; (4) Assess scope for using fungicides to ameliorate deleterious effects of host resistance responses. Significant yield costs associated with genetic resistance to septoria tritici blotch, yellow rust and brown rust were identified. Yield penalties were found to range between 0.3 – 1 t/ha depending on the resistance gene/QTL and genetic background of the variety.

Significant yield costs could be quantified by the measurement of yield, healthy area duration (HAD) of the crop canopy and pre-anthesis radiation use efficiency (RUE). Lr37, a ‘defeated’ brown rust resistance gene, exhibited yield losses in the absence of disease in three genetic backgrounds. Lines containing three septoria resistance QTL did not exhibit significantly greater yield losses than lines containing a single QTL, suggesting that ‘stacking’ of septoria resistance QTL within a variety may not increase yield costs. Not all resistance genes or QTL tested exhibited deleterious yield effects. It should therefore be possible to prioritise resistance genes in breeding programmes by selecting high productivity in the presence and absence of disease. A significant decrease in stomatal conductance and yield was associated with several Lr brown rust resistance genes (inc. Lr37) tested in the Thatcher background, in the presence or absence of pathogen challenge. In addition, metabolic analysis identified a number of Lr genes, including Lr37, as being associated with significant metabolic changes even in the absence of pathogen challenge. Detection of changes in host metabolism (from which changes in stomatal conductance may result in some cases) could prove a useful pre-breeding technique to screen for resistance genes which are at risk of exhibiting deleterious yield effects. There was indirect evidence from field trials suggesting that certain fungicides may ameliorate physiological costs of resistance responses, where spore germination is reduced.