Modeling and estimation of pollen-mediated gene flow at the landscape scale

Abstract

Abstract Undesired pollen-mediated gene flow (PMGF) in large-scale agriculture has received considerable attention for their ability to affect biological invasions. However, it has been not yet sufficiently understood how landscape structure and large-scale anthropogenic dispersal influence the gene spreading process. To explore the possible gene flow from crop species to wild relatives, we performed a literature review in ecosystem PMGF research for four major transgenic crops (maize, rice, wheat, oilseed rape). We found that whether pollen flow results induce a high cross-pollination rate mainly depends on the types of genetically modified (GM) species and cropping practice. Interestingly, the increase in the plot size of a pollen donor can lead to the increase in PMGF under ideal field conditions. However, while in a much larger landscape, it rapidly drops due to the exponential decrease in pollen density and the relatively low out-crossing rates. In large-scale field surveys, seeds of those plants with dormancy capability amplify the occurrence of transgene invasion and can remain alive in the seedbank for several years. Furthermore, in comparison with the distance-based methods, more and more spatio-temporal simulation models have proposed that a higher up-scaling percentage of GM crops is acceptable in a coexistence area of GM and non-GM crops. This highlights that the fate of (trans)genes should be emphasized in designing coexistence approaches. These results carry implications for gene dispersal management and indicate areas for future work, such as how to incorporate the response functions of different pollination factors into simulating models.