Field studies showed that there were same trends of the abundance change of coleoptera predators in rice ecosystem in 3 rice-cropping seasons in Minghui 63 paddies. 3 species of coleoptera predators, M. discolor, Ophionea indica (Thunberg), Paederus fuscipes Curtis were main predators which feed on soft-bodied insects in whole rice growth period. In Sanya rice ecosystem, adult M. discolor was the most dominant coleoptera predator in whole rice growth period especially in blooming stage in 3 rice seasons (Fig 1) (S1 File). Adults M. discolor were 19.85±2.20, 28.35±0.89, 24.55±1.73 in each replicate in early-blooming stage in Spring, Summer and Autumn rice seasons, 2–6 fold the quantities in booting, tillering and filling stage of 3 rice-cropping seasons.

Over 75% percent adult M. discolor survived the 26 days pollen feeding period, and the survival rates of males were slightly higher than that of females (Table 1) (S2 File). For both sexes, survival did not differ between the Bt and non-Bt pollen treatments. Similarly, the pre-oviposition period, total fecundity and female and male dry weight were not affected by the Bt pollen treatment (Table 1). Additionally, the daily fecundity of females was not significantly affected by feeding on Cry1Ac/Cry1Ab -expressing rice pollen (RM-ANOVA; F_1,76 = 0.12, P = 0.74) (Fig 2) (S3 File).

The ELISA measurements showed that the mean (±SE) concentration of Cry1Ab, Cry1Ac protein in the pollen was 15.26 ± 0.02μg /g and 11.92±0.01μg /g dry weight. No Cry1Ac or Cry1Ab protein was detected in any sample of non-Bt rice Minghui63 pollen.

When the experiment was terminated after 26 days, ELISA results showed that adults fed on Bt rice pollen contained Cry1Ac/Cry1Ab concentrations with a mean (±SE) of 2.09±0.01μg/g Cry1Ac and 1.55±0.02 μg/g Cry1Ab dry weight in female adults, 0.36±0.11μg/g Cry1Ac and 0.25±0.01μg/g Cry1Ab dry weight in male adults. No Cry toxin was detected in adults fed on non-Bt rice pollen.

Over 70% percent adult M. discolor fed BPH reared on Bt rice Huahui1 or Minghui 63 plants survived the 26 days period (Table 2) (S4 File). For both sexes, survival did not differ between BPH from Bt and non-Bt treatments. The pre-oviposition period, total fecundity and female and male dry weight were not different between 2 treatments (Table 2) (S4 File). The daily fecundity of females was not significantly affected by feeding on BPH from Huahui 1 rice (RM-ANOVA; F_1,64 = 0.67, P = 0.42) (Fig 3) (S5 File).

The ELISA measurements showed that the mean (±SE) concentration of Cry1Ab, Cry1Ac protein in the BPH reared on Huahui 1 was 3.23 ± 0.01μg /g and 2.46±0.01μg /g dry weight. No Cry1Ac or Cry1Ab protein was detected in any sample reared on non-Bt rice.

When the experiment was terminated after 26 days, ELISA results showed that female adults fed on BPH reared on Bt rice contained Cry1Ac/Cry1Ab concentrations with a mean (±SE) of 0.15±0.00μg/g Cry1Ab and 0.11±0.00μg/g Cry1Ac dry weight. Most male adults samples were detected no Cry1Ac/Cry1Ab concentration. No Cry toxin was detected in adults fed on BPH reared on non-Bt rice.

The survival rates of adult M. discolor were above 65% and did not differ between adults feeding on pure artificial diet or any Cry toxin treatment (Chi-square test, P> 0.05) (Table 3) (S6 File). While consumption of diet containing E-64 caused a decline in adult survival by more than 10%, this difference also was not significant (χ2 = 1.36, P = 0.24) (Table 3) (S6 File). No significant differences were detected between the Cry1Ac/Cry1Ab treatment and the control for the pre-oviposition period, total fecundity and adult dry weight (all P > 0.05) (S6 File). In contrast, the pre-oviposition period was significantly prolonged (Mann—Whitney U test; U = 58347.5, p = 0), the total fecundity was significantly reduced (Dunnett test; P < 0.05) when M. discolor adults were fed with the E-64-containing diet compared with those fed with pure artificial diet (Table 3) (S6 File). Likewise daily fecundity did not differ between the Cry1Ab, Cry1Ac or Cry1Ab/ Cry1Ac with control treatments (P>0.05), while E-64 caused a significant reduction in daily fecundity (RM-ANOVA; F_1,49 = 54.364, P < 0.01) (Fig 4) (S7 File).

After 26 days diet feeding, the ingestion of Cry1Ac/Cry1Ab by adult M. discolor was confirmed by ELISA. Similar to Bt rice pollen feeding, the females contained much higher mean (±SE) concentrations of the protein (Cry1Ab 2.55 ± 0.02 μg/g, Cry1Ac 1.89± 0.03 μg/g dry weight) than males (Cry1Ab 0.41 ± 0.01μg/g and Cry1Ac 0.31 ± 0.02μg/g dry weight). No Bt protein was detected in adults fed pure artificial diet.

No specific permits were required for the described studies.

Transgenic Bt Huahui 1 and its non-transgenic counterpart Minghui 63 were evaluated.

M. discolor and Nilaparvata lugens(Stal) were collected from rice fields in the experimental farm of Sanya Science&Technology Academy for crop winter multiplication during May 2012 and since then maintained in the laboratory without introductions of field-collected insects. Eggs collected from the colony were kept in a climatic chamber at 16:8 h (L: D), 28±1°C, and 75 ± 5%. After hatching, the larvae were kept individually in glass tubes (2 cm diameter, 9 cm long). 5–15 Drosophila adults were supplied daily to the tubes. After 6 generation newly emerged (<24 h after emergence) adults M. discolor were used for the experiments.

N. lugens maintained in rice Yuzhenxiang seedlings in the greenhouse in the experimental base of Tropical biological technology insititute in Wenchang couty. After more than 20 generation N. lugens were transferred to Huahui1 and Minghui 63 seedlings, 3rd instar N. lugens from Huahui1 and Minghui 63 were used for the experiments.

Cry1Ab toxin and Cry1Ac toxin (99%) used in this study was provided by EnvirologixInc. E-64(advanced pure) was provided by the Amresco Company.

There are three rice-cropping seasons according to sowing date in the studied area. The non-transgenic control Minghui 63 were sowed in early December in 2012, early April in 2013 and early July in 2013, and transplanted 1 month later. Field insect sampling by the D-vac method in minghui63 field were conducted at the experimental farm of Sanya Science&Technology Academy for crop winter multiplication in Hainan Province, China in 2013. Five samples were taken from each plot. For taking a sample, a plastic bin covered 20x20 cm² was placed over one hill and all arthropods inside the bin were carefully collected with a blower-vac machine sampler. There were 3 replicates, and each plot was 400 m². Rice seedlings were transplanted by hand at a density of 3 seedlings per hill. Spacing between rows was 20 cm and hills within rows were spaced 15 cm apart.

Field insect sampling were conducted in rice tillering, booting, early-blooming, blooming and filling stage with five point sampling method in every rice seasons. Each sampling site contained 9 rice hills. All treatments were managed by normal cultural practices except for pesticide sprays. Collected insects were put in 80% alcohol, and Coleoptera predators were counted in laboratory.

Pollen was obtained from greenhouse-grown rice of the Bt rice lines Huahui 1 and the nontransgenic line Minghui 63 in the Wenchang experimental base of Tropical biological technology insititute.

In the morning, inflorescences were cut from plants before anthesis. The obtained pollens was placed in sealed plastic vials and kept at—20°C until use. In the laboratory, anthers enclosed within glumes were striped off by scissors.

The experiment was conducted in the climatic carbine with photoperiod 16:8 h (L: D), 28±1°C, and 75 ± 5% RH.

For each treatment, new emergenced adults were transferred to a Petri dish (10.0 cm diameter, 2.6 cm tall). After 3 days, adults were paired and placed in glass tubes (2cm diameter, 9cm long) containing sufficient Bt or non-Bt rice pollens or BPH reared on Huahui1 or Minghui 63 and a water-saturated cotton ball with 5% leechee honey. Thereafter, daily fecundity and adult survival were recorded every 6 h. After 26 days, dry male and female adults were weighed.

When egg clutches were observed, adults M. discolor as well as the rice pollens/BPH were transferred to other clear glass tubes, and the numbers of egg clutches were recorded.

The purified toxin bioassay was carried out as described above. The experimental system was used to assess the toxicity of Cry1Ac and Cry1Ab on M. discolor. Five diet treatments were tested: 1) artificial diet (negative control) containing of litchi honey, egg yolk powder (in proportions 6:3); 2) artificial diet containing Cry1Ab at 150 mg/g weight of diet; 3) artificial diet containing Cry1Ac at 120 mg/g dry weight of diet; 4) artificial diet containing Cry1Ab at 150 mg/g and Cry1Ac at 120 mg/g dry weight of diet; 5) artificial diet containing E-64 at 25 mg/g fresh weight of diet (positive control). Diets were replaced every day. The experiment was conducted with 30–33 pairs neonate M. discolor adults per treatment. The insects were observed every 6 h and their survival and development were recorded.

The concentrations of Cry1Ab and Cry1Ac in rice pollen, adult M. discolor and BPH were measured by enzyme-linked immunosorbent assays (ELISA) using the Cry1Ab/Cry1Ac detection kits from Envirologix (Portland Maine, USA). Prior to analysis, insects were washed in phosphate buffered saline with Tween-20 (PBST) buffer (provided in the kit) to remove any Bt toxin from their outer surface.

After adding PBST to the samples at a ratio of at least 1:10 (mg sample:ml buffer) in 1.5 ml centrifuge tubes, the samples were fully ground by hand using a plastic pestle. After centrifugation and appropriate dilution of the supernatants, ELISA was performed according to the kit’s instructions. The measured OD values were calibrated to a range of concentrations of Cry1Ab and Cry1Ac made from purified toxin solution.

For the pollen bioassay, statistical analyses were conducted between the Bt pollen and non-Bt pollen treatments. For the BPH bioassay, statistical analyses were conducted between BPH from Bt rice Huahui 1 and non-Bt rice Minghui 63 treatments. In the purified toxin experiment, comparisons were made between the Cry1Ac, Cry1Ab, Cry1Ac /Cry1Ab or E-64 treatment with control (pure artificial diet). For both experiments, Chi-square tests were carried out to compare adult survival (females and males separately). Mann—Whitney U-tests were used to compare the preoviposition. Data on the total fecundity and adult dry weight were compared using Student’s t-tests in the pollen feeding bioassay and by Dunnett tests in the purified protein bioassay. Daily fecundity data were analyzed using repeated measures (RM) ANOVA.

All statistical analyses were conducted using the software package SPSS (version 16; SPSS, Inc., Chicago, IL).