Institute of Phytomedicine, University of Hohenheim, D-70593, Stuttgart, Germany

Background and objectives
Fusarium head blight of wheat, caused predominantly by Fusarium graminearum and F. culmorum, is a destructive disease in the humid and semi-humid wheat-growing areas throughout the world. The disease not only results in a significant yield reduction, but also contaminates grains with mycotoxins. Fusarium toxins, including deoxynivalenol (DON), 3-acetyl-DON, 15-acetyl-DON and nivalenol, have been found in infected wheat heads [1]. However, the toxin distribution in infected wheat spikes during disease development is so far unknown. Since Fusarium toxin may be involved in pathogenesis, the localization of the toxin in wheat spike at the early stages of infection has been investigated by immunogold labelling.

Materials and methods
The spikelets of a susceptible cultivar, Agent, were inoculated at anthesis with a macroconidial suspension of Fusarium culmorum and lemma, rachis and kernels were sampled 2, 4 and 6 days after inoculation (d.a.i.). To determine if Fusarium toxin could be translocated in the infected spike, the middle spikelets of spikes were inoculated and the rachis three spikelets above and below the inoculated spikelets were sampled at 14 d.a.i. Sample processing for electron microscopy was carried out according to [2]. For immunogold labelling, after blocking with TBS containing gelatin and goat normal serum, thin sections were incubated with anti-3-acetyl-DON antiserum (kindly provided by Dr E. Usleber, University of Munich) diluted 1:100 in TBS, and then with goat-anti-rabbit-1gG-gold diluted 1:40 in TBS. For control, the thin sections were incubated as above, except that the antiserum was preincubated with 3-acetyl-DON. Following staining, the sections were examined with electron microscopy.

Results and conclusions
Observations from electron microscopy showed that hyphae of the pathogen grew on the surface of the lemma at 2 d.a.i. Gold labelling occurred in the cell wall, cytoplasm and vacuoles of the hyphae, but the organelles sulch as mitochondria of the hyphae were free of significant labelling. At this time, the labelling was found in the host cell wall, cytoplasm and chloroplasts of the lemma. The results indicate that the pathogen had produced its toxin already before invading the host and the toxin diffused into the host cell. At 4 d.a.i., hyphae were observed only in intercellular spaces, and host cells were affected. Electron-dense material appeared between the host cell wall and plasmalemma. More gold labelling was found in the host cell wall and cytoplasm. Host cells then collapsed, and cytoplasm and chloroplasts had degenerated br 6 d.a.i. The pathogen could be found inter- and intracellularly in the lemma and rachis. More dense labelling was detected in vessel walls of the xylem. The observations above demonstrate that the toxin was associated with pathological changes of host tissue during infection, which suggest that the toxin is one of the pathogenic factors of the pathogen. No hyphae were found in the rachis three spikelets above or below the inoculated spikelets at 14 d.a.i., however, gold labelling occurred in the vessel wall of the xylem in the upper rachis, as well as in the sieve cytoplasm, sieve walls and sieve tubes of the phloem in the lower rachis. These results demonstrate that the toxin in the infected wheat spike could move in both the xylem and phloem.

1. Atavassov Z, Nakamura C, Mori N et al., 1994. Canadian Journal of Botany 72, 161-167.
2. Sossountzov L, Sotta B, Matdiney R et al., 1986. Planta 168, 471-481.