Genome evolution in an agricultural pest following adoption of transgenic crops

Replacement of synthetic insecticides with transgenic crops for pest management has been both economically and environmentally beneficial. These benefits have often eroded as pests evolved resistance to transgenic crops, but a broad understanding of the timing and complexity of adaptive changes which lead to field-evolved resistance in pest species is lacking. Wild populations of Helicoverpa zea, a major lepidopteran crop pest and the target of transgenic Cry toxin-expressing cotton and corn, have recently evolved widespread, damaging levels of resistance. Here, we quantified patterns of genomic change in wild H. zea collected between 2002 and 2017 when adoption rates of Cry-expressing crops expanded in North America. Using a combination of genomic approaches, we identified significant temporal changes in allele frequency throughout the genomes of field-collected H. zea. Many of these changes occurred concurrently with increasingly damaging levels of resistance to Cry toxins between 2012 and 2016, in a pattern consistent with polygenic selection. Surprisingly, none of the eleven previously described Cry resistance genes showed signatures of selection in wild H. zea. Furthermore, we observed evidence of a very strong selective sweep in one region of the H. zea genome, yet this strongest change was not additively associated with Cry resistance. This first, whole genome analysis of field-collected specimens to study evolution of Cry resistance demonstrates the potential and need for a more holistic approach to examining pest adaptation to changing agricultural practices. Significance StatementEvolution of pest resistance to management approaches in agricultural landscapes is common and results in economic losses. Early detection of pest resistance evolution prior to significant crop damage would benefit the agricultural community. It has been hypothesized that new genomic approaches could track molecular signals of emerging resistance problems and trigger efforts to pre-empt widespread damage. We tested this hypothesis by quantifying genomic changes in the pest Helicoverpa zea over a 15 year period concurrent with commercialization and subsequent loss of efficacy of transgenic Bt-expressing crops. Our results demonstrate the complex nature of evolution in agricultural ecosystems and provide insight into the potential for and pitfalls associated with use of genomic approaches for resistance monitoring. We discuss approaches for improvement.