Effect of plant water deficit on the deoxynivalenol concentration in Fusarium-infected maize kernels

In current climate change scenarios, mean air temperatures and summer droughts are expected to increase over the long-term average in large parts of Europe. These changes will strongly affect the growth and health of cultivated plants. In a field experiment in 2009 and 2010 in rain-out shelters, the consequences of plant water availability under three water regimes on the severity of Fusarium ear rot, deoxynivalenol (DON) contamination and yield of maize were investigated. Water was provided exclusively to the plants by a mobile sprinkler system installed in the rain-out shelter. Three maize cultivars were supplied with 50, 75, and 125% of the long-term average monthly precipitation of the experimental site. In 2009, Fusarium graminearum-infected oat kernels were placed on the soil surrounding maize plants to promote ear infection, whereas in 2010, the maize silks were directly inoculated with suspensions of F. graminearum conidia. Heavy drought stress in the 50% water regime was reflected in the average ear yield of the three maize cultivars of 75 dt ha-1 compared with 192 dt ha-1 at 125% water. In this comparison, the DON concentrations in the kernels were fivefold higher at 50% water than at the high water regime: 380 compared with 75 μg DON kg-1 DM. In 2010, the drought stress symptoms were less pronounced than in 2009, and a much lower ear yield loss from 128 to 108 dt DM ha-1 was observed with decreasing water supply. The DON contamination of the kernels was at a higher level than in 2009, but a similar upward trend from 330 μg kg -1 DM at 125% water supply to about 3.5-fold higher DON levels at 75 and 50% water supply was observed. These different yield responses presumably resulted from variable climatic conditions at the experimental site in the 2 years. The results of this study suggest that the risk of DON contamination of maize kernels increases when plants are grown under conditions of long-term water deficit.