Does refuge spillover affect arthropod food webs associated with Bt maize?

Abstract

BACKGROUND: A high dose/refuge combination is the main tactic recommended for mitigating resistance selection of target herbivore species in crops expressing insecticidal proteins of the bacterium Bacillus thuringiensis (i.e. Bt proteins). The tactic consists of the simultaneous use of Bt crops expressing high levels of the Bt protein associated with neighboring areas of refuge of the same non-Bt crop species. Nonetheless, the approach faces controversy regarding its effectiveness and scale of adoption, at least in some regions. One concern focuses on its potential impact on the arthropod community, including its short-term and spatially dependent impact considering the likely biota spillover effect between Bt and non-Bt neighboring areas. Thus, the eventual spillover of Bt maize targeted and non-targeted arthropods was surveyed along transects extending from the refuge border to the center of the Bt maize area. RESULTS: Arthropods were collected throughout the maize vegetative and reproductive stages. A total of 85 arthropod species were collected, but their richness and abundance did not vary with distance from the refuge. By contrast, cultivation season played a significant role in distinguishing the arthropod communities. Refuge distance from the sampling point within Bt-fields did not significantly affect the food web metrics, unlike season, which affected the number of nodes integrating each food web. Winter maize cultivation exhibited higher arthropod diversity and combined values of species numeric abundance and biomass at each trophic level. CONCLUSIONS: No arthropod spillover was evident between the refuge edge and Bt maize, adding further controversy to the tactic currently subjected to lower usage in the region with a disputed cost-benefit relationship, because not even the target and its interdependent species were affected.

Arthropod food webs associated with cotton: Does Bt cotton mediate community stress?

“…Bt and non‐Bt), cultivation period and replicate (i.e. field) was used for the purpose (Haro et al., 2018; Zuim et al., 2021).…”

“…Only the species collected present in more than 5% of the samples were considered for the analyses to prevent rare species from interfering with the results. The log‐transformed data from these species were initially subjected to a selective process to detect the species contributing the most for the observed variance in the study using the STEPDISC procedure with STEPWISE selection (SLENTRY = 0.15) on SAS (SAS Institute; Badji et al., 2004; Zuim et al., 2021). The species selection was based on (i) the significance level on an F test from covariance analyses, where the chosen species perform as covariates and the treatments are the dependent variables; and (ii) a squared partial correlation to predict the treatments from the species, controlling for the effects of the model's previously selected species.…”

“…boll weevil, whitefly and cotton aphid), which are not targets of the technology (Blanco et al., 2016). Thus, we were not expecting a significant change in the cotton‐associated community of arthropods either, unless the prevalence of cotton caterpillars was dominant, an unlike expectation in the region contrasting with that of maize (Guedes et al., 2017; Sujii et al., 2013; Zuim et al., 2021).…”

“…Some studies did use more robust and high‐resolution approaches, including multivariate techniques for the ordination and recognition of the impact on species assemblages (Torres & Ruberson, 2005). Nonetheless, assessments exploring direct and indirect species interactions indicating potential community‐level stress are lacking, except for recent related efforts on maize (Zuim et al., 2021). This is important not only because Bt cotton may shift the dominance of prevailing pest species, as suspected with the US cotton plant bugs (Eisenring et al., 2019), but also because the suppression of cotton pest caterpillars may lead to bottom‐up effects on associated natural enemies in the area (Guedes et al., 2016, 2017).…”

“…Food web analysis using trophic networks is a powerful and promising tool to assess the community‐level impact of a range of stressors, including Bt crops, particularly when using species and genus identification of the collected organisms to increase resolution (Guedes et al., 2017; Szénási et al., 2014; Zuim et al., 2021). The use of such a tool is timely, as the current stacking of Bt proteins in cotton may provide even more drastic effect on the target caterpillars, while mitigating the evolution of their resistance to individual proteins (Liu et al., 2019; Siegfried & Jurat‐Fuentes, 2016; Xiao & Wu, 2019).…”