Iron (Fe) is the most important transition metal in all living organisms, while it is also highly toxic to cells by generating hydroxyl radical with the Fenton reaction. This Fe-dependent redox toxicity can be used to prevent the spread of pathogen infection, but also inhibits plant growth, reduces rice yield in flooded conditions or acerbates the deficiency of other nutrients such as phosphate deficiency. In agriculture, serious symptoms of Fe toxicity have been recognized in rice for more than half a century. It is estimated that 19% of the total area for rice production in Africa harbors a potential risk for Fe toxicity. Therefore, engineering Fe toxicity tolerant rice varieties would be highly desirable. However, more than twenty attempts over the past two decades were not successful in identifying genes responsible for the variation of Fe toxicity tolerance in rice using linkage and association mappings. Thus, unlike several important molecular components involved in Fe uptake, transport and homeostasis have been identified in past decades, the genes and mechanisms are responsible for plant tolerance to this Fe-dependent redox toxicity are largely unknown in both plants and animals.

We identified S-nitrosoglutathione-reductase (GSNOR) variants underlying a major quantitative locus for root tolerance to Fe-toxicity using genome-wide association studies in Arabidopsis thaliana that are under much less artificial selection (domestication) compared with crops. These GSNOR variants act largely through transcript level regulation. GSNOR maintains root meristem activity under high Fe levels and prevents cell death via inhibiting Fe-dependent nitrosative and oxidative cytotoxicity, which reveals that nitric oxide is also required to generating Fe-dependent redox toxicity. Moreover, we demonstrated that GSNOR is required for root tolerance to Fe-toxicity throughout higher plants such as legumes and monocots. Our findings not only shed light the underlying quantitative genetic and molecular mechanisms of plant adaptation to Fe-dependent redox toxicity, but also expose GSNOR as a prime target for genetic editing or marker-assisted breeding to develop crop varieties tolerant to Fe-dependent toxicity and help to increase rice productivity in regions with potential Fe toxicity.