Antifeedant and ovicidal activities of ginsenosides against Asian corn borer, Ostrinia furnacalis (Guenee)

Introduction Ginsenosides, including protopanaxdiol (PPD) and protopanaxtriol (PPT) type ginsenosides, have been identified as natural insecticidess. This study aimed to investigate the antifeedant and ovicidal activities of total ginsenosides, protopanaxdiol saponins (PDS) and protopanaxtriol saponins (PTS) against Asian corn borer, O. furnacalis (Guenee). Methods and results O. furnacalis egg masses (> 40 eggs) at 0-, 1- and 2-day-old were dipped into ginsenosides and egg hatchability was significantly inhibited by total ginsenosides, PDS, and PTS in dose and egg-age dependent manners. 100 mg/ml PDS had the strongest ovicidal activity against 0- (80.58 ± 0.95%), 1- (71.48 ± 5.70%), and 2-day-old eggs (64.31 ± 3.20%). In no-choice and choice feeding tests, we observed that the 3rd instar larvae consumed decreased area of leaves treated with ginsenosides, and the antifeedant activities of total ginsenosides, PDS, and PTS against the 3rd instar larvae were time and dose-dependent, with higher activities at 48 h. 100 mg/ml PDS had relative higher antifeedant activity (88.39 ± 3.43% in no-choice and 80.9±4.36% in choice) than total ginsenosides and PTS at all time intervals, except at 48 h in no-choice test. In further experiments, we found PPD ginsenosides (Rb1, Rb2, Rc, and Rd) had relative higher time and dose dependent antifeedant activities than PPT ginsenosides (Re and Rg1). Conclusions Our results suggested the insecticidal action of total ginsenosides, PDS, and PTS on O. furnacalis. All ginsenosides, especially PDS, showed antifeedant and ovicidal activities against O. furnacalis.

Our previous study showed that total ginsenosides from the stems and leaves (GSLS) of P. ginseng showed significant antifeedant activity against Plutella xylostella (Linnaeus) [21]. We observed that the administration of GSLS extracted from P. ginseng significantly reducted the activity of acetylcholine esterase (AChE), carboxylesterase (CarE) and glutathione S-transferase (GST), while increased mixed-functional oxidase (MFO) activity in P. xylostella, suggesting the decreasing detoxication capacity in P. xylostella. Other studies also showed the obvious efficient of ginsenosides against pest infestation by influencing feeding behavior of Mythimna separata (M. separata Walker) Larvae [22]and Acyrthosiphon pisum (A. pisum pea aphid) [14], as well as digestive enzymes in M. separata larvae [23] and A. pisum [14]. However, there is less information on the activity of pest ginsenosides against infestation from Ostrinia furnacalis (O. furnacalis).
O. furnacalis is a widely distributed insect serviously threatening the production of maize in Asia [24,25]. The control of O. furnacalis is very difficult work for agricultural management in Aisa. This study was designed to investigate the antifeedant and ovicidal activities of ginsenosides against O. furnacalis. The influence of series concentrations of ginsenosides were tested on egg hatchability and feeding behavior (in no-choice and choice situations). The antifeedant and ovicidal activities of total ginsenosides from the leaf and stem of P. ginseng, PDS, and PTS against O. furnacalis eggs or the 3 rd instar larvae were investigated. This study would provide us with new insights into the insecticidal action of different ginsenosides against O. furnacalis.

Insects
The O. furnacalis eggs were obtained from the Institute of Biocontrol, Jilin Agricultural University, China, and maintained in an intelligent artificial climate chamber (PRS-20, Ningbo Saifu Laboratory Instrument Factory, Ningbo, China) at 27 ± 1˚C with a relative humidity of 70 ± 5% and a light cycle of 14 h day: 10 h night. Insects were fed with artificial diets for 5 generations with undisturbed mating behavior. The indoor populations of O. furnacalis and their eggs were then subjected to treatments with ginsenosides.

Ovicidal activity bioassay
For ovicidal bioassay, 10 egg masses (� 40 eggs in each egg mass) at each age (0-, 1-, and 2-day-old) of the indoor populations of O. furnacalis were immersed into solutions with different concentrations of ginsenosides (0, 5, 25, 50 and 100 mg/ml) for 10 s. After eggs masses were air dried, 10 eggs were placed per 9cm petri dish each containing a moistened filter paper, with 10 replications per treatment. Dishes were incubated in a chamber as described above. The unhatched eggs (unbroken eggs) were counted daily using a Leica binocular microscope (Leica Microsystems, Germany) until hatching ceased. Control experiments were conducted by dipping egg masses into distilled water for 10 s. Each experiment was repeated for 10 times. The egg hatchability (eggH) was calculated as: eggH(%) = (hatched eggs)×100/total eggs. The ovicidal activity of ginsenoside was repressed by the adjusted inhibition rate of egg hatchability, which was calculated as: Adjusted inhibition rate of egg hatchability(%) = (eggH control -eggH test )×100/eggH control .

Feeding bioassay
Leaf disc choice and no-choice methods were used to study the larvicidal activity. Fresh maize leaves were collected from 'Xianyu 335' at jointing stage. Leaves were imidiately stored at 4˚C before experiments. Leaf discs (15 mm in diameter) were dipped into solutions with different concentrations of ginsenosides (0, 5, 25, 50 and 100 mg/ml) for 10 s. Leaf discs dipped into distilled water were used as negative control. The leaf discs were placed on filter papers and air dried. In choice bioassay, 3 test discs and 3 negative control discs dipped into the same solutions were put into one petri dish (9 cm in diameter) with interval [21]. In no choice bioassay, 6 test discs were put into one dish, and dishes with 6 control discs were set as negative controls [21]. Each experiment was replicated for 10 times. The 3 rd instar larvae were fasted for 4 h and then evenly placed onto the empty positions of dishes (n = 6 in each dish) with leaf discs. All dishes were maintained in a chamber as described above for 24 h, 48 h and 72 h. The remaining area (ReA) of individual leaf disc was determined using a leaf area meter (LI-3000; LI-COR, Lincoln, NE, USA). Then, the antifeedant activity of ginsenoside against O. furnacalis larvae for choice bioassay was computed as: antifeedant activity(%) = (ReA control -ReA test )×100/(ReA control + ReA test ), and that for no-choice assay was computed as: antifeedant activity(%) = (ReA control -ReA test )×100/(ReA control ).
It has been reported that Rb1, Rb2, Rc, Re and Rg1 are major compositions (80%) of total ginsenosides [15]. As we described above, Re, Rd, Rg1, Rb3 and Rb1 are major compositions (~45%) of GSLS from P. ginseng; PDS was mainly composted of PPD type ginsenosides including Rd, Rc, Rb2, Rb1 and Rb3 [15,17]; and PTS was essentially consisted of Re, Rg1, and Rg2 (accounting for~55%) [17]. To investigate the reasons causing the difference in antifeedant activity among GSLS, PDS, and PTS, we fed the 3 rd instar larvae with leaf discs treated with different active ginsenoside monomers, including PPD ginsenosides (Rb1, Rb2, Rc, and Rd) and PPT type ginsenosides (Re and Rg1). Experiments were performed as aforementioned for 10 replications in each condition.

Statistical analysis
All data were expressed as mean ± SD from 10 replications. Statistical analyses were performed using SPASS 22.0 software. Data were analyzed using one-way ANOVA followed by a post hoc Dunnet's test for comparion. Median lethal concentration (LC 50 ) and antifeedant concentration (AFC 50 ) of ginsenosides were calculated using regression analysis. P < 0.05 was set as significant different.
Our analysis results suggested that there were differences in compositions of GSLS, PDS and PTS, which might suggest the different activities of GSLS, PDS and PTS against O. furnacalis.

Ovicidal activity of GSLS, PDS and PTS against the corn borer eggs
We firstly analyzed the inhibitory effect of GSLS, PDS and PTS against the hatchability of O. furnacalis eggs. Fig 3 shows that all ginsenosides significantly inhibited egg hatchability in a dose-dependent manner. The detail data were listed in S1 Table. After being treated with GSLS (5~100 mg/ml), the hatchability of eggs at 0-, 1-and 2-day-old was in a range of 71.70%2 5.98%, 78.67%~32.69% and 38.61%~82.86%, respectively, which was lower than 96.86%9 8.38% of control (Fig 3A-3C and S1 Table). The hatchability of 0-day-old eggs treated by PDS was significantly lower than eggs treated by GSLS and PTS at the same concentration, while that of the 1-and 2-day-old eggs treated by PDS was insignificantly lower than GSLS and PTS (Fig 3A-3C). On contrast, the ovicidal activity of ginsenosides was gradually increased with ginsenoside concentrations (S1 Table and Fig 3D-3F). The ovicidal activity of 100 mg/ml PDS was the highest (80.58±0.95%, 71.48±5.70% and 64.31±3.20%) against O. furnacalis eggs (at 0-, 1-and 2-day-old). We found the ovicidal activity of PDS for 0-, 1-, and 2-day-old eggs was higher than that of GSLS and PTS in each concentration (Fig 3D-3F). These results were consistent with our previous report showing the activity of GSLS against oviposition of Pieris rapae [2]. Our results confirmed the ovicidal activity of GSLS, PDS and PTS against the hatch of O. furnacalis eggs, suggesting that GSLS, PDS and PTS were potential inhibitors of O. furnacalis egg hatch.
Regression analysis showed that PDS exhibited the LC 50 values of 23.01 mg/ml, 41.20 mg/ ml, and 59.75 mg/ml for ovicidal activity against eggs at 0-, 1-, and 2-day-old, respectively, which was lower than LC 50 of GSLS (48.08 mg/ml, 58.78 mg/ml and 71.63 mg/ml) and LC 50 of PTS (65.56 mg/ml, 67.29 mg/ml, and 73.99 mg/ml; Table 2) based on confidence intervals. These results suggested that GSLS, PDS and PTS all inhibited the O. furnacalis egg hatchability in dose-dependent manners, and egg age affected the ovicidal activity of ginsenosides. PDS had relative higher ovicidal activity against O. furnacalis eggs at all ages (0~2-day-old) compared with PTS and GSLS from the leaf and stem of P. ginseng, based on LC 50 values and nonoverlapping confidence intervals. Larvae that fed on the leaf discs treated with highest ginsenoside concentrations (100 mg/ml) consumed the minimum leaf area, suggesting the highest antifeedant activity of 100 mg/ml ginsenosides against the 3 rd instar larvae (Fig 4 and S2 and S3 Tables). In addition, PDS showed higher antifeedant activity than others in the same concentration (S2 and S3 Tables).
Choice bioassay revealed PDS had the highest antifeedant activity against the 3 rd instar larvae in all experiments (Fig 4D-4F). 100 mg/ml PDS showed the highest antifeedant activity of 70.88 ± 2.34%, 80.9 ± 4.36%, and 47.88 ± 4.71% at 24 h, 48 h, and 72 h, respectively, and those were significantly higher than GSLS and PTS (p <0.05, S3 Table). In no-choice and choice assays, antifeedant activities of ginsenosides were relative lower at 72 h than those at 24 h and 48 h. These results of no-choice and choice bioassays agreed with our previous studies showing the inhibitory activity of GSLS against insect feeding behavior [1,2,21]. Previous studies had determined the antifeedant activity of GSLS against the 3 rd instar larvae of Pieris rapae [2], the 4 th instar larvae of M. separata [1] and Plutella xylostella (Linnaeus) [21].
We calculated the choice and no-choice AFC 50 of GSLS, PDS and PTS at all time intervals. Regression analysis of no-choice bioassay showed PDS exhibited the lowest AFC 50 value of 30.86 mg/ml, 7.06 mg/ml, and 23.61 mg/ml at 24 h, 48 h and 72 h, respectively (Table 3). AFC 50 values of PTS in choice and no-choice tests were higher than those of GSLS and PDS based on LC 50 values.
Taken together, these results revealed that GSLS, PDS and PTS all showed antifeedant activities against the 3 rd instar larvae of O. furnacalis, and PDS had the highest antifeedant activity. Fresh maize leaf treated with higher ginsenoside concentrations exhibited higher antifeedant activity against the 3rd instar larvae of O. furnacalis, suggesting the potential strategies of using ginsenosides for O. furnacalis management.   Table). All  ginsenoside monomers showed relative higher antifeedant activity against O. furnacalis larvae at 48 h, followed by 72 h and then 24 h (Fig 5 and S4 Table). In comparison among monomers, we found Re and Rg1 monomers in the same concentrations showed relative lower antifeedant  Tables,  respectively. https://doi.org/10.1371/journal.pone.0211905.g004 activities than Rb1, Rb2, Rc and Rd, and the antifeedant activities of Rb1, Rb2, Rc and Rd were at the same level (Fig 5 and S4 Table). Regression analysis showed the AFC 50 value of Rb1, Rb2, Rc, Rd, Re and Rg1 was 0.32 mg/ml, 0.33 mg/ml, 0.45 mg/ml, 0.37 mg/ml, 0.81 mg/ml and 0.64 mg/ml at 48 h, respectively (Fig 6). Previous reports had shown that compound K and Rg3 (PPD type) have various pharmacological activities, including anti-tumorigenesis, anti-inflammatory, neuroprotective and antiaging [6,[27][28][29]. Ginsenosides Rb1, Rb2, Re and Rd could be converted into pharmacologically active components PPD type ginsenosides compound K and Rg3 during biotransformation [28,29]. Oh et al reported that PPT and PPD showed comparable anxiolytic effect on rats  Table. https://doi.org/10.1371/journal.pone.0211905.g005

Antifeedant activity of different ginsenoside monomers against the 3 rd instar larvae of O. furnacalis
Insecticidal action of ginsenosides and the latter showed stronger anti-inflammatory effect than the former [20]. These studies suggested that the diversity biological in activities of ginsenoside are dependent on the structures and biotransformation pathways. These might explain the differences in the ovicidal and antifeedant activities among GSLS, PDS and PTS ginsenosides against O. furnacalis.

Conclusion
In conclusion, our present study determined the insecticidal action of GSLS, PDS and PTS against O. furnacalis in laboratory conditions. We confirmed that the ovicidal activities of ginsenosides GSLS, PDS and PTS against O. furnacalis eggs were dose and egg-age dependent. PDS showed the highest ovicidal activity among the three types (GSLS, PDS, and PTS) against O. furnacalis eggs at 0~2-day-old. No-choice and choice antifeeding tests showed the timeand dose-dependent antifeedant activities of GSLS, PDS and PTS against the 3 rd instar larvae of O. furnacalis. Moreover, PDS exhibited relative stronger activity against feeding behavior of the 3 rd instar larvae than PTS and GSLS. The minor but interesting differences in antifeedant activities among ginsenoside monomers (PPD type ginsenosides Rb1, Rb2, Rc, and Rd; and PPT type ginsenosides Re and Rg1) showed the ginsenoside structures might be associated with the insecticidal activities of ginsenosides. Accordingly, we confirmed the potential of using ginsenosides, especially PDS, as the treatment strategies for controlling O. furnacalis. The potential should be validated in field experiments. Moreover, the uncovering of molecular mechanism in insects in response to different ginsenosides might be helpful for understanding the insecticidal action of GSLS, PDS and PTS against O. furnacalis.