Extent of pollen-mediated gene flow and seed longevity in switchgrass (Panicum virgatum L.): Implications for biosafety procedures

New switchgrass (Panicum virgatum L.) bioenergy cultivars are being bred through genetic engineering; however, baseline information is urgently needed to establish guidelines for small-scale field trials prior to commercialization. In this study, we documented the pattern of pollen-mediated gene flow and the extent of seed longevity in field experiments. To mimic crop-to-wild, pollen-mediated gene flow, we planted wild recipient switchgrass ramets at various distances away from cultivar donor ramets at two sites in Ohio. Percent hybridization at each distance was estimated from seed set on recipient ramets, which were self-incompatible clones. The pattern of gene flow was best described by negative exponential models, and the minimum isolation distance for a 0.01% gene flow threshold was predicted to be 69 m and 109 m away from the pollen source at the two sites. To investigate seed longevity, we buried seeds of six cultivars and ten wild biotypes in Ohio and Iowa in 2011. A subset of the seeds were exhumed, germinated, and tested for dormancy over three years. Cultivars lost seed viability and dormancy significantly sooner than wild biotypes at both locations in the first year, and most biotypes lost dormancy by the second year. Cultivar seeds buried in the cooler, drier Iowa site had an overall greater longevity than those buried in Ohio. Our findings suggest that substantial amounts of pollen-mediated gene flow could occur in the immediate vicinity of switchgrass pollen sources, and current switchgrass cultivars are unlikely to persist in the seed bank for more than three years.