Controlling plant disease has been a struggle for mankind since the advent of agriculture. Studies of plant immune mechanisms have led to strategies of engineering resistant crops through ectopic transcription of plants’ own defence genes, such as the master immune regulatory gene NPR11. However, enhanced resistance obtained through such strategies is often associated with significant penalties to fitness2, making the resulting products undesirable for agricultural applications. To remedy this problem, we sought more stringent mechanisms of expressing defence proteins. Based on our latest finding that translation of key immune regulators, such as TBF13, is rapidly and transiently induced upon pathogen challenge (accompanying manuscript), we developed “TBF1-cassette” consisting of not only the immune-inducible promoter but also two pathogen-responsive upstream open reading frames (uORFsTBF1) of the TBF1 gene. We demonstrate that inclusion of the uORFsTBF1-mediated translational control over the production of snc1 (an autoactivated immune receptor) in Arabidopsis (At) and AtNPR1 in rice enables us to engineer broad-spectrum disease resistance without compromising plant fitness in the laboratory or in the field. This broadly applicable new strategy may lead to reduced use of pesticides and lightening of selective pressure for resistant pathogens.

BackgroundTransgenic Bt rice line T2A-1 expresses a synthesized cry2A gene that shows high resistance to Lepidoptera pests, including Cnaphalocrocis medinalis (Guenée) (Lepidoptera: Pyralidae). Plant volatile orientation cues and the physical characteristics of the leaf surface play key roles in host location or host-plant acceptance of phytophagous insects. These volatile compounds and physical traits may become altered in Bt rice and it is not known whether this influences the behavior of C. medinalis when searching for oviposition sites.ResultsThe results of electronic nose analysis showed that the Radar map of Bt rice cultivars was analogous to the non- Bt rice cultivars at each growing stage. PCA analysis was able to partly discriminate between some of the Bt vs. non-Bt rice sensors, but could not to separate Bt cultivars from non-Bt cultivars. The total ion chromatogram between Bt and non-Bt rice cultivars at the seedling, booting and tillering stages were similar and 25 main compounds were identified by GC-MS. For most compounds, there was no significant difference in compound quantities between Bt and non-Bt rice cultivars at equivalent growth stages. The densities of the tubercle papicles and the trichomes on the upper and lower surfaces were statistically equal in Bt and non-Bt rice. The target pest, C. medinalis, was attracted to host rice plants, but it could not distinguish between the transgenic and the isogenic rice lines.ConclusionsThere were no significant differences between the Bt rice line, T2A-1 and the non-Bt rice for volatiles produced or in its physical characteristics and there were no negative impacts on C. medinalis oviposition behavior. These results add to the mounting evidence that Bt rice has no negative impact on the target insect oviposition behavior.