Establishment and optimization of a regionally applicable maize gene-flow model

Abstract

Because of the rapid development of transgenic maize, the potential effect of transgene flow on seed purity has become a major concern in public and scientific communities. Setting a proper isolation distance in field experiments and seed production is a possible solution to meet seed-quality standards and ensure adventitious contamination of products is below a specific threshold. By using a Gaussian plume model as basis and data recorded by meteorological stations as input, we have established a simple regionally applicable maize gene-flow model for prediction of the maximum threshold distances (MTD) at which gene-flow frequency is equal to or lower than a threshold value of 1 or 0.1 % (MTD1%, MTD0.1%). After optimization of the model variables, simulated outcrossing rate was a good fit to data obtained from field experiments (y = 1.156x, R (2) = 0.8913, n = 30, P < P 0.01). In the process of model calibration, it was found that only 15.82 % of the total amount of the pollen released by each plant participated in the dispersal process. The variable "a" for genetic pollen competitiveness between donor and recipient was introduced into our model, for the "Zinuo18" and "Su608" used, "a" was 17.47. Finally, the model was successfully used in the spring maize-growing region of Northeast China. The range of MTD1% and MTD0.1% in this region varied from 10 m to 49 m and from 17 m to 125 m, respectively.

High crop barrier reduces gene flow from transgenic to conventional maize in large fields

“…Langhof et al (2010) found that an isolation distance of 50 m in the Germany is enough to make cross-pollination rate below 0.9% (Langhof et al, 2010). With a Gaussian plume model, Hu et al (2014) predicted that the maximum threshold distances were 10-49 m at which gene-flow frequency is lower than 1% using 101 meteorological data in the maize-growing region of Northeast China. Thus, isolation distance of 50 m is safe to meet the 0.9% threshold.…”

Model-based calculating tool for pollen-mediated gene flow frequencies in plants

“…In addition, PMGF frequencies estimated from a limited number of field experiments may not represent PMGF frequencies of a particular plant species under diverse environmental conditions. As a comparison, mathematical modelling can provide relatively quick estimation of PMGF frequencies for different plant species under diverse environmental conditions (Yao et al 2008; Rong et al 2010; Hu et al 2014). Undoubtedly, mathematical modelling provides a powerful complement of the field-experiment-based method to estimate PMGF frequencies, provided that the model can accurately simulate PMGF (Rong et al 2010; Wang et al 2016).…”

“…Many PMGF models have been established (Loos et al 2003; Klein et al 2006; Yao et al 2008; Wang and Yang 2010; Hu et al 2014). However, some of these models essentially simulate the variation patterns/dynamics of PMGF or describe parameters that affect PMGF frequencies (Loos et al 2003; Klein et al 2006).…”

“…Such features make these models difficult to be applied directly to predict PMGF frequencies under specific situations. Furthermore, many PMGF models only include sophisticated equations or functions (Yao et al 2008; Wang and Yang 2010; Hu et al 2014), which puzzles the users who are not familiar with the theories and skills of modelling. It is difficult for the users to directly use these models to estimate PMGF frequencies.…”

“…It is difficult for the users to directly use these models to estimate PMGF frequencies. For example, some PMGF models only included mathematical equations that were generated from two-dimensional Gaussian distribution and double integral to describe PMGF patterns (Yao et al 2008; Hu et al 2014), challenging for common users. The above limitation hinders the application of mathematical modelling for the prediction of PMGF frequencies.…”