Utilization of DNA sequences for identifying Fusarium species isolated from rice

Abstract

We attempted the identification of Fusarium species by a DNA sequence analysis to practically validate the utility of a molecular approach for fungal identification and reveal its limitations. We sequenced three regions, the 5' end of the large-subunit (D2 region) and the internal transcribed spacer 1 and 2 (ITS1 and ITS2) regions, in the rRNA genes. The DNA sequences of 38 Fusarium strains isolated from unpolished rice harvested in Japan were analyzed and compared for similarity to entries in the GenBank. Based on this comparison, it was estimated that all these three regions, as a minimum, must be compared with the database to identify Fusarium fungi at the species level. According to the sequence differences in the three regions, the 38 isolates were classified into 13 groups. Out of the 13 groups, 6 groups (20 isolates in total) were able to be identified as the definite species based only on the sequence data. For the other 6 groups (17 isolates in total), the candidate species were restricted on the basis of the sequence similarity. The restriction of candidate species rendered it easy to identify the Fusarium isolates at the species level with the aid of their morphologies. Only one isolate could not be identified. It was verified in this study that a DNA sequence comparison with the GenBank database was useful and reliable for the identification of the Fusarium species.