No abstract

Little is known about the potential cumulative long-term effects of transgenic crops on nontarget organisms. In the present laboratory study, the potential cumulative effects of transgenic Bacillus thuringiensis (Bt) rice on parasitoids in successive generations were observed for an egg parasitoid, Anagrus nilaparvatae parasitizing eggs of Nilaparvata lugens (Stål) (Hemiptera: Delphacidae) feeding on Bt rice. Enzyme-linked immunosorbent assay test confirmed that Cry1Ab insecticidal protein could be detected in newly eclosed parasitoid adults. However, no significant effect on the fecundity of Anagrus nilaparvatae Pang et Wang (Hymenoptera: Mymaridae) was observed between Bt and non-Bt rice. Developmental times of both genders of A. nilaparvatae parasitizing host eggs laid in Bt (KMD1 and KMD2) rice lines were significantly prolonged from first generation to second generation, but not always prolonged from third generation to 11th generation as compared with the control rice line. Furthermore, the sex ratio of A. nilaparvatae progeny from the first generation to 11th generation in three rice lines was not significantly different. In general, our results suggested that the effect of Bt rice on this parasitoid could be negligible.

Summary1. Rice is one of the most important crops in the world. Several transgenic varieties of rice have been developed, and some have recently entered pre-production trials. One concern with genetically modified (GM) crops is transgene escape, but prior studies suggest this risk is low for rice because it is self-pollinated and the dispersal of pollen by wind is limited. 2. Little is known about the impact of pollen transport by insects. We characterized the insects visiting rice plants during anthesis and considered the effects of insect pollination on gene flow. 3. We conducted a 2-year nationwide survey in China and identified more than 510 insect species that visited rice flowers. Honeybees, hoverflies and several other species carried large amounts of pollen. The European honeybee Apis mellifera visited rice flowers regularly with daily foraging activity peaking between 12.00 and 13.00 h. 4. We monitored 20 European honeybee colonies located 100-1000 metres away from rice fields in mixed agricultural landscapes and found the honeybees carried viable pollen at least 500 m away from the rice pollen source. 5. We used four GM rice lines as pollen donors, their non-GM parental varieties as pollen recipients and the European honeybee as the pollinator in field-cage experiments to assess whether honeybees increase the frequency of gene flow in rice. Results from screening over 1Á5 million germinated offspring seeds over a 3-year study period showed that honeybees significantly increase transgene flow in rice. 6. Synthesis and applications. Our results indicate that a remarkably high diversity of insects visit rice flowers in China and that hundreds of species including honeybees carry large amounts of rice pollen. European honeybees carry viable pollen over long distances, forage on rice flowers regularly and increase the frequency of transgene flow. Insects mediate gene flow in rice more than previously assumed, and this should be taken into consideration during the ecological risk assessment of transgene flow in self-pollinated and/or anemophilous crops.

The effects of three transgenic Bacillus thuringiensis (Bt) rice lines, KMD1, KMD2, and G8-7, on biological parameters and population dynamics of nontarget insect, Rhopalosiphum maidis (Fitch) (Homoptera: Aphididae), were investigated in the laboratory and field. No significant differences were found between Bt and non-Bt rice lines for aphid survival. The developmental time of R. maidis that fed on KMD1 and KMD2 did not differ significantly from those of the individuals feeding on the parental variety Xiushui11, but significantly prolonged developmental time was observed on G8-7 as compared with its parental variety Xiushui110. Aphid fecundity was significantly higher on Bt than on parental rice. A 2-yr field survey indicated that Bt rice did not significantly affect the population dynamics of R. maidis in comparison with non-Bt rice. Additionally, guttation droplets of Bt rice and aphids feeding on Bt rice were analyzed for presence of Cry1Ab using ELISA. No Cry1Ab protein was found in aphid adults feeding on Bt rice lines both in the laboratory and field. By using the guttation droplets from the top of rice seedlings, we designed a novel method to collect phloem sap, and found that relatively low concentrations were detected in the guttation droplets from Bt rice lines. In conclusion, although the Bt rice lines tested in this study stimulate the fecundity of R. maidis, the aphid population density did not increase in Bt rice fields.