Abstract
AbstractWeeds, a main threat to agricultural productivity worldwide, are mostly controlled by herbicides. To minimize herbicide usage by targeting it to weedy areas, we developed a new image-based methodology for robust weed detection that relies on manipulating the crop plant’s leaf hue, without affecting crop fitness. We generated transgenic tobacco (Nicotiana tabacum Xanthi) lines overexpressing anthocyanin pigment as a traceable marker that differentiates transgenes from the surrounding weeds at an early stage. Transformation with the anthocyanin VlmybA1-2 gene produced purple-colored leaves. Subsequent gene silencing with vector pTRV2:VlmybA1-2 significantly reduced anthocyanin pigments in tobacco leaves 40 days after agroinfiltration, with a concomitant reduction in VlmybA1-2 transcript levels. Purple hue faded gradually, and there were no fitness costs in terms of plant height or leaf number in the silenced vs. non-silenced tobacco transgenes. These results could lead to a new sustainable weed-control method that will alleviate weed-related ecological, agricultural and economic issues.
Publisher
Cold Spring Harbor Laboratory