Characterization of the rice blast resistance gene Pik cloned from Kanto51

To study similar, but distinct, plant disease resistance (R) specificities exhibited by allelic genes at the rice blast resistance locus Pik/Pikm, we cloned the Pik gene from rice cultivar Kanto51 and compared its molecular features with those of Pikm and of another Pik gene cloned from cv. Kusabue. Like Pikm, Pik is composed of two adjacent NBS-LRR (nucleotide-binding site, leucine-rich repeat) genes: the first gene, Pik1-KA, and the second gene, Pik2-KA. Pik from Kanto51 and Pik from Kusabue were not identical; although the predicted protein sequences of the second genes were identical, the sequences differed by three amino acids within the NBS domain of the first genes. The Pik proteins from Kanto51 and Kusabue differed from Pikm in eight and seven amino acids, respectively. Most of these substituted amino acids were within the coiled-coil (CC) and NBS domains encoded by the first gene. Of these substitutions, all within the CC domain were conserved between the two Pik proteins, whereas all within the NBS domain differed between them. Comparison of the two Pik proteins and Pikm suggests the importance of the CC domain in determining the resistance specificities of Pik and Pikm. This feature contrasts with that of most allelic or homologous NBS-LRR genes characterized to date, in which the major specificity determinant is believed to lie in the highly diverged LRR domain. In addition, our study revealed high evolutionary flexibility in the genome at the Pik locus, which may be relevant to the generation of new R specificities at this locus.