Quantitative exposure assessment for confinement of maize biogenic systems
Is Version Of
The development of transgenic crops as production platforms for biogenic agents will largely depend on the success of efforts to confine the genes and their expressed proteins in field environments. We have used quantitative exposure assessment to evaluate how management practices affect materials escape due to outcrossing by pollen flow or grain loss during harvest operations. Specifically, we study the use of maize to produce biogenic agents within field-confined systems. Decision trees representing simplified schemes of fully conforming (designed to comply with current regulatory standards for field confined trials), partially conforming, and non-conforming management practices were developed. Exemplifying assumptions and published data for conformance and material fate probabilities were used in Monte Carlo simulations to forecast materials escape by pollen outcrossing and harvest operations from a 1 ha source field. Deterministic analyses showed fully conforming confinement management restricted materials loss to low levels (for this example, outcrossing produced <1 in 106 kernels in receptor fields). The corresponding high-end (90th percentile) probabilistic result was 16- and 4333-fold higher (relative to deterministic outcrossing = 1) for outcrossing and harvest loss, respectively. For partially conforming practice, high-end outcrossing ranged from 100- to >15 000-fold over the base result in receptor fields, and harvest loss was >10 000-fold over the base result. For non-conforming practice, high-end outcrossing produced >15 000-fold greater kernels in receptor fields and high-end harvest loss was at least 19 000-fold greater. Deterministic estimates of off-field loss by machine transfer are as much as 30 000-fold higher for non-conforming operations relative to the base case of pollen outcrossing. Better knowledge of failure frequencies for confinement management practices, improved physical models of materials flows, refined analysis of confinement loss probabilities using quantitative tools, and decision analysis to improve and audit management system performance are all needed to extend understanding of confinement integrity beyond the exemplifying case used here.
This article is from Environmental Biosafety Research 3 (2004): 183–196, doi:10.1051/ebr:2005004. Posted with permission.