Effects of cultivar, agronomic practices, geographic location, and meteorological conditions on the composition of selected Fusarium species on wheat heads

Abstract

Wheat (Triticum aestivum) heads showing typical fusarium head blight symptoms were sampled two to four weeks after anthesis in 2004, 2005, and 2006. Samples (n = 916) were taken from the susceptible wheat cultivar 'Ritmo' and the moderately susceptible 'Buteo' and 'Dekan', from fungicide-free control plots at 6 to 13 locations per year between the North and Baltic seas. The presence of four Fusarium species (Gibberella zeae, Fusarium culmorum, Fusarium poae, and Fusarium avenaceum) in these samples was surveyed using (i) genus and (ii) species-specific primers in PCR assays. Higher frequencies of G. zeae and F. avenaceum were observed in 'Ritmo' compared with 'Buteo', and lower frequencies of F. culmorum were observed in 'Dekan' compared with 'Buteo'. Tillage (ploughing vs. other tillage methods) and previous crops (wheat vs. dicotyledonous crops) had no significant effect on the Fusarium species composition ('Ritmo', P ≥ 0.07). Average leaf wetness between May and the middle of July, as well as during the 10 days flanking the approximate date of anthesis, was positively correlated to relative frequencies of G. zeae (R > 0.44, P ≤ 0.04), but negatively correlated to relative detection frequencies of F. culmorum (R < –0.47, P ≤ 0.02). Temperature, precipitation, and relative humidity during the same periods of time did not influence the detection frequencies observed for any of the species. Overall, the most frequently detected species was G. zeae (41%), with decreasing relative frequencies from the southeast to the northwest of the area surveyed. Fusarium poae, F. culmorum, and F. avenaceum were detected in 12%, 9%, and 13% of the samples, respectively. The presence of one, two, three, or four species co-inhabiting wheat heads was encountered at frequencies of 57%, 19%, 2%, and 0%, respectively. Gibberella zeae and F. culmorum were mainly detected alone, whereas F. poae and F. avenaceum were mainly detected in combination with other species. Relative frequencies of PCR amplicons specific for the genus Fusarium were positively correlated to the deoxynivalenol (DON) content of the grain (R = 0.685, P ≤ 0.0001). The correlation between the percent of samples containing F. poae and the DON content was negative (R = –0.45, P = 0.036). The results of this study indicate that tillage and previous crop may affect the severity of Fusarium epidemics, but they do not change the species composition significantly. In contrast, wheat genotype and leaf wetness affected the spectrum of Fusarium species on wheat heads in a selective manner, and, hence, these factors might also be expected to have an impact on the spectrum of mycotoxins in wheat grain.

Keywords:

• host-pathogen interaction
• mycotoxin
• species-environment interaction
• Triticum aestivum