Estimating Gene Flow in Wheat and Jointed Goatgrass: A Progress Report
In the era of genetically engineered (GE) crops, pollen-mediated gene flow is a concern for two reasons: (1) if the introduced genes spread to standard varieties of the crop, it may cause marketing problems for the conventional varieties, especially for export markets; and (2) if the genes are transferred to wild species, such as jointed goatgrass, negative environmental effects may result. Although GE wheat cultivars are not expected to be released in the near future, they will likely be introduced at some point. Therefore, to provide relevant information for regulatory agencies and growers, we have undertaken a three-year project to estimate gene flow in commercial-scale wheat fields, with funding from USDA’s Biotechnology Risk Assessment Research Grants Program. Our project takes advantage of the Colorado release of ‘Above’, a Clearfield (imazamox herbicide tolerant) winter wheat variety. By sampling seeds of non-Clearfield varieties in fields adjacent to ‘Above’ fields and testing for their tolerance to Beyond (imazamox) herbicide, we can estimate the amount of cross-pollination that has taken place. The assumption is that herbicide tolerance in the standard varieties will be due almost entirely to genes obtained from the nearby source of ‘Above’ pollen.
We have now completed evaluation of 129 samples collected from 17 Eastern Colorado locations in 2003. Another 268 samples were collected from 20 locations during the 2004 harvest season (Fig. 1).
|
|---|
| Figure 1: Locations of wheat and jointed goatgrass samples collected in 2003 and 2004. |
Herbicide tolerance of seeds collected in 2003 was evaluated in both greenhouse and field plantings; only the field data are presented here. Approximately 15,000 seeds per sample were planted at CSU’s Agricultural Research, Development, and Education Center near Fort Collins in October, 2003. In early to mid-spring, 2004, the plots were sprayed twice with recommended rates of Beyond herbicide. Our objective was to kill plants that did not contain herbicide tolerance genes, but to allow survival of those plants that had received a single dose of the tolerance gene from ‘Above’ pollen. In early May, when control plants were in the boot stage, the numbers of survivors and dead plants were counted and the percentage of survivors calculated.
Table 1. Percent cross-pollination as determined by herbicide tolerance for a subset of locations and varieties sampled in 2003.
Sample location (county) |
Variety |
Direction from Above |
Distance from Above, ft |
Percent herbicide tolerance |
Baca |
Prairie Red |
N |
0.5 |
1.20 |
N |
40 |
0.15 |
||
N |
81 |
0.08 |
||
N |
120 |
0.00 |
||
S |
0.5 |
1.08 |
||
S |
40 |
0.06 |
||
S |
81 |
0.04 |
||
KitCarson 1 |
Jagger |
NE |
0.75 |
2.58 |
NE |
20 |
0.94 |
||
NE |
40 |
0.29 |
||
NE |
120 |
0.32 |
||
W |
0.75 |
5.31 |
||
W |
20 |
3.34 |
||
W |
40 |
3.13 |
||
W |
120 |
0.50 |
||
Kit Carson 2 |
Ike |
NE |
0.75 |
0.26 |
NE |
20 |
0.11 |
||
NE |
40 |
0.17 |
||
NE |
120 |
0.11 |
||
W |
0.75 |
0.07 |
||
W |
20 |
0.09 |
||
W |
40 |
0.01 |
||
W |
120 |
0.03 |
||
Sedgwick |
Akron |
E |
1 |
0.77 |
E |
20 |
0.02 |
||
E |
40 |
0.11 |
||
Ankor |
W |
1 |
0.07 |
|
W |
20 |
0.01 |
||
W |
40 |
0.01 |
||
Weld |
Yuma |
N |
3 |
0.08 |
N |
20 |
0.02 |
||
N |
40 |
0.01 |
||
N |
60 |
0.02 |
||
N |
90 |
0.01 |
||
N |
120 |
0.01 |
||
Preliminary results from a subset of locations and varieties are presented in Table 1. As expected, the highest rates of cross-pollination occurred at sampling positions closest to ‘Above’ and in varieties with heading dates similar to ‘Above’, such as ‘Prairie Red’ and ‘Jagger.’ Cross-pollination rates dropped off rapidly with distance, but we detected low levels of herbicide tolerance as far as 120 feet from ‘Above.’ The direction relative to ‘Above’ in which samples were collected had an important influence on degree of cross-pollination in some locations. Only a slight difference was seen between north and south directions at a location in Baca County, but a major difference occurred between directions at two sites in Kit Carson County. These differences are most likely due to the direction of prevailing winds during the morning hours when most wheat pollen is shed. Although we assume that the herbicide tolerance observed in this study is due to cross-pollination, other explanations, such as seed impurities, cannot be completely ruled out. Please bear in mind that the data presented here represent a single season. We will evaluate samples for two more seasons before making more general conclusions about the level of gene flow in eastern Colorado wheat fields.
This project would not have been possible without the excellent cooperation we received from growers and extension agents during the past two seasons. We hope to continue those collaborations during the 2004-05 season.
   |