Molecular detection of Diaporthe phaseolorum and Phomopsis longicolla from soybean seeds

Species-specific detection of Diaporthe phaseolorum and Phomopsis longicolla from soybean seeds was accomplished using polymerase chain reaction-restriction fragment length polymorphism (PCR-RFLP) and TaqMan chemistry. To use these detection systems, fungal DNA was released from soybean seed coats using an ultrasonic processor to break the cells. DNA fragment lengths ranged from 200 to 1,200 base pairs (bp), with the majority of fragments <500 bp. Based on DNA sequences of the internal transcribed spacer (ITS) regions of ribosomal DNA, three TaqMan primer/probe sets were designed. Primer/probe set PL-5 amplified a 96-bp fragment within the ITS1 region of P. longicolla, D. phaseolorum var. caulivora, D. phaseolorum var. meridionalis, and D. phaseolorum var. sojae. Set PL-3 amplified a 86-bp DNA fragment within the ITS2 region of P. longicolla. Set DPC-3 amplified a 151-bp DNA fragment within the ITS2 region of D. phaseolorum var. caulivora. TaqMan primer/probe sets were able to detect as little as 0.15 fg (four copies) of plasmid DNA. When using PCR-RFLP for Diaporthe and Phomopsis detection, the sensitivity was as low as 100 pg of pure DNA. Among 13 soybean seed lots from Italy and the United States, the total Diaporthe and Phomopsis detected using a traditional seed-plating technique ranged from 0 to 32%. P. longicolla was most prevalent, followed by D. phaseolorum var. sojae. D. phaseolorum var. caulivora, which only occurred in 0.5% of the Italian seed lots, was not detected in the U.S. seed lots. D. phaseolorum var. meridionaIis was not detected in either the U.S. or Italian seed lots. Using TaqMan primer/probe set PL-3, the frequency of P. longicolla was 18% in seed lot 13, similar to the frequency obtained from PCR-RFLP and potato dextrose agar plating detection. The frequencies of D. phaseolorum and P. longicolla in each seed lot obtained by the different detection methods were comparable with respect to total infection and individual species detection. However, TaqMan detection provided the fastest results of all the methods tested.