Caused by Sclerotinia sclerotiorum, white mold is an important seed-transmitted disease of soybean (Glycine max). Incubation-based methods available for the detection and quantification of seed-borne inoculum such as the blotter test, paper roll and Neon-S assay are time-consuming, laborious, and not always sensitive. In this study, we developed and evaluated a molecular assay for the detection of S. sclerotiorum in soybean seeds using a species-specific PCR (polymerase chain reaction) primer set and seed soaking (without DNA extraction) for up to 72 h. The PCR products were amplified in all the samples infected with the pathogen, but not in the other samples of plant material or the other seed-borne fungi DNA. The minimum amount of DNA detected was 10 pg, or one artificially infested seed in a 400-seed sample (0.25 % fungal incidence) and one naturally infected seed in a 300-seed sample (0.33 % incidence). The PCR-based assay was rapid (< 9 h), did not require DNA extraction and was very sensitive.

O objetivo deste trabalho foi avaliar a eficiência dos métodos de detecção de Sclerotinia sclerotiorum em sementes de soja. Foram realizados experimentos durante três safras, tendo-se utilizado amostras de sementes de soja contaminadas artificial e naturalmente, com incidência conhecida da doença nas amostras e no campo. Os métodos de incubação utilizados foram: papel de filtro, rolo de papel e meio Neon-S. Observou-se aumento da detecção do patógeno nas sementes infectadas artificialmente, à proporção que aumentou a incidência na amostra. Os métodos podem não detectar o patógeno S. sclerotiorum em sementes de soja naturalmente contaminadas. Os níveis de incidência do patógeno em sementes de soja interferem na sensibilidade dos métodos de detecção, e não há relação entre a incidência no campo e na semente colhida.

The objective of this work was to evaluate the efficiency of detection methods of Sclerotinia sclerotiorum on soybean seeds. Experiments were carried out during three crop years, using samples of soybean seeds artificially and naturally contaminated, with known pathogen incidence on samples and on the field. The incubation methods used were: filter paper, paper roll, and Neon-S medium. An increase in pathogen detection on artificially contaminated seeds was observed, as the incidence in the sample increased. The methods may not detect the pathogen S. sclerotiorum on naturally contaminated soybean seeds. The levels of pathogen incidence on soybean seeds interfere in the sensitivity of the detection methods, and there is no relationship between the incidence on the field and on the harvested seed.