RESUMO
Herbicide mixtures are used to increase the spectrum of weed control and to manage weeds with target-site resistance to some herbicides. However, the effect of mixtures on the evolution of herbicide resistance caused by enhanced metabolism is unknown. This study evaluated the effect of a fenoxaprop-p-ethyl and imazethapyr mixture on the evolution of herbicide resistance in Echinochloa crus-galli using recurrent selection at sublethal doses. The progeny from second generations selected with the mixture had lower control than parental plants or the unselected progeny. GR50 increased 1.6- and 2.6-fold after two selection cycles with the mixture in susceptible (POP1-S) and imazethapyr-resistant (POP2-IR) biotypes, respectively. There was evidence that recurrent selection with this sublethal mixture had the potential to evolve cross-resistance to diclofop, cyhalofop, sethoxydim, and quinclorac. Mixture selection did not cause increased relative expression for a set of analyzed genes (CYP71AK2, CYP72A122, CYP72A258, CYP81A12, CYP81A14, CYP81A21, CYP81A22, and GST1). Fenoxaprop, rather than imazethapyr, is the main contributor to the decreased control in the progenies after recurrent selection with the mixture in low doses. This is the first study reporting the effect of a herbicide mixture at low doses on herbicide resistance evolution. The lack of control using the mixture may result in decreased herbicide sensitivity of the weed progenies. Using mixtures may select important detoxifying genes that have the potential to metabolize herbicides in patterns that cannot currently be predicted. The use of fully recommended herbicide rates in herbicide mixtures is recommended to reduce the risk of this type of resistance evolution.