Rice phenolamindes reduce the survival of female adults of the white-backed planthopper Sogatella furcifera

In response to infestation by herbivores, rice plants rapidly biosynthesize defense compounds by activating a series of defense-related pathways. However, which defensive compounds in rice are effective against herbivores remains largely unknown. We found that the infestation of white-backed planthopper (WBPH) Sogatella furcifera gravid females significantly increased levels of jasmonic acid (JA), jasmonoyl-isoleucine (JA-Ile) and H2O2, and reduced the level of ethylene in rice; levels of 11 of the tested 12 phenolamides (PAs) were subsequently enhanced. In contrast, WBPH nymph infestation had no effect on levels of JA, JA-Ile, ethylene and H2O2 in rice, and enhanced levels of only 2 of 12 PAs. Moreover, infestation by brown planthopper Nilaparvata lugens gravid females also affected the production of these PAs differently. Bioassays revealed that 4 PAs – N-feruloylputrescine, N-feruloyltyramine, feruloylagmatine and N1,N10-diferuloylspermidine – were toxic to newly emerged WBPH female adults. Our results suggest that WBPH- or BPH-induced biosynthesis of PAs in rice seems to be shaped primarily by the specific profile of defense-related signals elicited by the herbivore and that PAs play a role in conferring the resistance to WBPH on rice.

N-p-coumaroylputrescine (CouPut) or FerPut in rice leaves increased following the infestation of herbivores, including by the brown planthopper (BPH) Nilaparvata lugens and that both of these PAs reduced the survival rate of BPH 15 . However, thus far, only a few PAs have been reported to play an important role in conferring plant resistance to herbivores.
Rice, Oryza sativa, one of the most important staple crops worldwide, is severely infested by herbivores 16 . Infestation of rice by insect herbivores, including white-backed planthopper (WBPH) Sogatella furcifera and BPH, changes levels of a variety of defense-related phytohormones, including JA, JA-Ile, SA and ethylene [17][18][19] . Changes in these compounds, in turn, cause rice plants to produce defense responses, such as the increased activity of trypsin protease inhibitors, the production of PAs and the release of volatiles; these responses enhance the direct and/or indirect resistance of rice to herbivores [18][19][20] . BPH infestation has been reported to reduce the resistance of rice to WBPH and therefore improve the performance of WBPH on rice plants 21 . Moreover, as mentioned above, BPH female adult infestation induced the production of CouPut and FerPut 15 . Yet, whether infestation of WBPH induces the production of PAs in rice, whether and which PAs are defensive against WBPH and whether the suppression of BPH infestation on the resistance of rice to WBPH is related to PA levels remain unknown.
In this study, we found that infestation by WBPH gravid females induced the production of JA, JA-Ile and H 2 O 2, and repressed the biosynthesis of ethylene, which subsequently enhanced levels of 11 of the tested 12 PAs; infestation by WBPH nymphs, on the other hand, did not alter levels of JA, JA-Ile, H 2 O 2 and ethylene, and elicited accumulation of only 2 of 12 PAs. Unlike WBPH gravid female infestation, BPH gravid female infestation induced the production of only 3 PAs while suppressing the biosynthesis of 4. Bioassays revealed that 4 PAs -FerPut, FerTyr, FerAgm and N1, N10-diferuloylspermidine (DiferSpe) -were toxic to WBPH female adults. The results demonstrate that (1) the influence of herbivore infestation on the biosynthesis of PAs in rice is probably related to the profile of signaling pathways that the herbivore elicited and (2) PAs confer the resistance to WBPH on rice.

Results
Infestation of WBPH gravid females but not of nymphs induces the biosynthesis of JA, JA-Ile and H 2 o 2 in rice. JA, JA-Ile and H 2 O 2 play vital roles in plant defense responses to herbivores 1-3 . Therefore, we investigated whether WBPH infestation induced the biosynthesis of these signals. We found that WBPH nymph infestation did not change levels of JA and JA-Ile in rice plants compared to in non-infested rice plants (Fig. 1). However, levels of JA and JA-Ile in rice plants at 3, 8 and 24 h after infestation by WBPH gravid females were 8.1-, 7.9-, 6.9-and 7.8-, 10.2-, 6.3-fold higher than levels of JA and JA-Ile in non-infested rice plants, respectively (Fig. 1a,b). Similarly, WBPH nymph infestation did not induce the production of H 2 O 2 in rice at 2 and Infestation of WBPH gravid females but not of nymphs suppresses the biosynthesis of ethylene in rice. Ethylene-mediated signaling also plays an important role in herbivore-induced plant defenses 1,3,18 .
In rice, ethylene signaling reportedly positively regulated the resistance of rice to the striped stem borer Chilo suppressalis but negatively mediated the resistance to BPH 18 . In this study, we observed that infestation of WBPH gravid females reduced the level of ethylene in rice: ethylene levels in rice plants infested by gravid WBPH females 24 and 48 h after infestation were 78.1% and 75.7% of those in non-infested rice plants (Fig. 2). Unlike the infestation of WBPH gravid females, the infestation of WBPH nymphs did not influence the production of ethylene in rice (Fig. 2). In summary, WBPH gravid female infestation suppressed the biosynthesis of ethylene in rice whereas WBHP nymph infestation did not.

BPH infestation influences the production of PAs in rice differently than WBPH infestation.
Levels of CouPut and FerPut in rice leaves were reported to increase when leaves were infested by BPH female adults 15 . Here we investigated the accumulation of PAs in rice leaf sheaths, the normal feeding and oviposition site of BPH on rice plants, after these sheaths had been infested with BPH gravid females for 3 days. The results showed that BPH gravid female infestation increased levels of 3 PAs -CinPut, DiferSpe and CouAgm-in rice leaf sheaths: levels of CinPut, DiferSpe and CouAgm in leaf sheaths of plants infested by BPH gravid females were 4.9, 7.2 and 24.3-fold, respectively, higher than those in leaf sheaths of non-infested plants (Fig. 4). Moreover, compared to levels in leaf sheaths of non-infested plants, levels of 4 PAs -CafPut, CouFerPut, DiferPut and FerPut Scientific RepoRtS | (2020) 10:5778 | https://doi.org/10.1038/s41598-020-62752-y www.nature.com/scientificreports www.nature.com/scientificreports/ -in leaf sheaths of plants infested by gravid BPH females decreased (Fig. 4). Levels of the 5 remaining PAs did not differ between BPH gravid female-infested plants and non-infested plants. BPH gravid female infestation clearly influences the production of PAs in rice differently from WBPH gravid female infestation.  Several PAs had a direct negative effect on the survival of WBPH female adults. As two PAs have been reported to reduce the survival of BPH female adults 15 , we investigated if some of the induced PAs also negatively influenced the survival of WBPH. Seven PAs -FerPut, FerTyr, FerAgm, CouAgm, CinPut, DiferSpe and CouPut -all of which were significantly induced by WBPH gravid female infestation, were chosen to be added to artificial diet to examine their direct influence on the performance of newly emerged WBPH female adults. Six d after the start of feeding, the survival rate of WBPH female adults fed on artificial diet with FerPut and FerTyr at concentrations of 100 μg/ml significantly decreased (by 60.5% and 62.8%, respectively), compared to the survival rate of WBPH female adults fed on artificial diet without PAs (Fig. 5a,b); the survival rate of WBPH female adults fed on artificial diet with FerAgm at a concentration of 100 μg/ml significantly decreased starting at 2 d after feeding, with a maximum decrease of 68.4% at 6 d (Fig. 5c). For DiferSpe, a concentration of 5 μg/ml caused significant reductions in the survival of WBPH female adults 6 d after feeding; when the concentration of DiferSpe rose to 100 μg/ml, the survival of WBPH decreased starting 2 d after feeding, reaching a minimum at 6 d (Fig. 5d). Three PAs -CinPut, CouAgm and CouPut -had no effect on the survival rate of WBPH female adults (Fig. 5e,f). In summary, WBPH gravid female-induced FerPut, FerTyr, FerAgm and DiferSpe were toxic to WBPH.

Discussion
In this study, we identified 12 PAs in rice and found that 11 of these PAs were induced by WBPH gravid female infestation, whereas, surprisingly, only 2 were induced by WBPH nymph infestation (Fig. 3). Bioassays revealed that 4 of the tested 7 PAs -FerPut, FerTyr, FerAgm and DiferSpe-reduced the survival of newly emerged WBPH female adults. Phytohormone analysis showed that WBPH gravid female infestation significantly increased levels of JA, JA-Ile and H 2 O 2, and reduced the level of ethylene in rice, whereas WBPH nymph infestation had no effect (Figs. 2, 3). These findings demonstrate that PAs play an important role in the resistance of rice to WBPH and that the infestation of rice by WBPH gravid females and nymphs induces levels of PAs, and these differ according to the defense-related signaling pathways that are elicited at the same time. www.nature.com/scientificreports www.nature.com/scientificreports/ PAs have been reported to be induced by pathogen infection 22,23 , herbivore infestation 15 and UV radiation 11 . Moreover, JA-and ethylene-mediated signaling pathways (though not the SA-mediated pathway) positively regulate the production of PAs 6,13 . We found that infestation by WBPH gravid females had a stronger influence on the biosynthesis of PAs than infestation by WBPH nymphs; this difference is probably due to different influences of WBPH gravid female infestation and WBPH nymph infestation on the production of defense-related signals, JA, JA-Ile, H 2 O 2 and ethylene. The reason why WBPH nymph infestation and WBPH gravid female infestation have different effects on levels of JA, JA-Ile, ethylene and H 2 O 2 is probably related to their different damage modes: nymphs just pierce and suck phloem sap, causing very little damage, whereas gravid females not only pierce and suck phloem sap but also lay eggs into leaf sheaths; oviposition (laying eggs into leaf sheaths) will result in more mechanical wounding compared to that caused by WBPH feeding 16 . Moreover, different effectors and/or elicitors derived from WBPH salivary glands/oral regurgitant and/or egg fluid may also contribute to this difference. It has been well documented that elicitors and/or effectors exist in both herbivore salivary glands/oral regurgitant and egg fluid 2,24-26 . In BPH, several elicitors and effectors from salivary glands have also been identified [27][28][29][30][31] . Interestingly, we also observed that WBPH and BPH gravid females induced PAs differently (Fig. 4): BPH infestation reduced levels of 4 PAs in rice sheaths, whereas these were induced by WBPH infestation, and one of these 4 PAs, FerPut, was toxic to WBPH; in addition, levels of 3 BPH-induced PAs -CinPut, DiferSpe and CouAgm -were generally lower than those induced by WBPH (Fig. 3). Although both WBPH gravid female infestation and BPH gravid female infestation have been reported to induce the production of JA, JA-Ile and H 2 O 2 , and to suppress the production of ethylene, the profile of these signals, including the level and timing of their emission, differs between the two insect species 32 . Hence, the herbivore-induced biosynthesis of PAs in rice seems to be shaped primarily by the specific profile of defense-related signals elicited by the herbivore.
PAs have been reported to enhance the resistance of plants to pathogens and herbivores via both reinforced cell walls, which reduce the digestibility of herbivores and invasion of pathogens, and direct toxicity for pathogens and herbivores 9,15,22,33 . We found that the survival rate of WBPH female adults fed on artificial diets complemented with certain concentrations of each of the following 4 PAs -FerPut (100 μg/mL), FerTyr (100 μg/mL), www.nature.com/scientificreports www.nature.com/scientificreports/ FerAgm (100 μg/mL) and DiferSpe (5 and 100 μg/mL) -decreased significantly. This suggests that some PAs are toxic to WBPH, just as they are to BPH adults 15 . However, the effect of PAs on different herbivores is not identical: CouPut, for example, has been found to reduce the survival rate of BPH adults 15 but had no negative influence on WBPH adults (Fig. 5). Although the levels of these toxic PAs determined here were lower than the levels used in bioassays, it should be emphasized that both BPH and WBPH feed phloem sap where levels of these PAs may be higher than in other tissues, a direction worthy for future analysis. It may be that levels of these PAs in plant phloem may be higher than in other tissues, a direction for future analysis. Moreover, Alamgir et al. 15 found that the level of FerPut in rice leaves under heavy BPH infestation could be up to 60 μg/g FM; this level is comparable to the level of FerPut that were effective we found in bioassays. Taken together, these findings demonstrate that PAs enhance the resistance of rice to WBPH.
Previous investigations have found that BPH infestation can improve the performance of WBPH, including enhancing WBPH survival rates and feeding amounts 21 . Considering that PAs are toxic to WBPH, the results observed in this study-namely, that fewer PAs were induced by BPH infestation than by WBPH infestation, and lower levels of PAs in were found in BPH-infested plants than in WBPH-infested plants -may be part of the explanation for the difference. Interestingly, we found that the effect of BPH infestation on the biosynthesis of PAs was different from the result reported in Alamgir et al. 15 . They showed that both CouPut and FerPut can be induced by BPH infestation, whereas we observed that BPH infestation reduced the level of FerPut and did not or only marginally induce the production of CouPut (Fig. 4). This discrepancy may be related to the different rice varieties used and rice tissues sampled: in the study of Alamgir et al. 15 , leaves from the japonica variety Nipponbare were sampled; we sampled leaf sheaths from the japonica variety XS110. Further research is needed to elucidate this issue.
In summary, rice PAs respond differently to WBPH gravid female infestation, WBPH nymph infestation and BPH gravid female infestation, probably because these herbivore infestations induced different profiles of defense-related signals. Like the negative effects of PAs on survival rates of BPH reported in Alamgir et al. 15 , we also found that PAs reduced survival rates of WBPH female adults. Given that PAs in rice can be induced by pathogens, such as blast fungus Magnaporthe oryzae and rice brown spot fungus B. oryzae 34 , and other insect pests, such as the lawn armyworm Spodoptera mauritia and the rice skipper Parnara guttata larvae 15 , and given that PAs are toxic to B. oryzae, Xoo 34 , WBPH and BPH, we expect them to play a vital role in interactions among rice, insect pests and pathogens. Elucidating the functions of PAs by using genetically modified techniques may decipher some of these complicated interactions.

Methods
Plant growth and insect rearing. The rice used in this study was a japonica type variety XS110. Plants were grown used the same method as described in Huangfu et al. 35 . Briefly, pre-germinated seeds of XS110 were sown in plastic bottles (diameter 8 cm, height 10 cm) in a greenhouse at 28 ± 2 °C and 14 h photophase, and after 10 days, the seedlings were transferred to 20-L hydroponic boxes with a rice nutrient solution 36 . Twenty-five days later, seedlings were transplanted to individual 500 mL hydroponic opaque plastic pots (diameter 8 cm, height 10 cm), each with one plant. Plants were used for experiments 4-5 days after transplanting.
WBPH and BPH colonies were originally obtained from rice fields in Hangzhou, China, and maintained on rice seedlings of TN1, a variety susceptible to WBPH and BPH, in a controlled climate room that was maintained at 26 ± 2 °C, 12 h photophase and 80% relative humidity. , and then analyzed with HPLC/mass spectrometry/mass spectrometry following the method described by Lu et al. 18 . Each plant treatment at each time interval was replicated five times. For ET analysis, plants infested with WBPH nymphs or gravid females and control plants were covered with sealed glass cylinders (diameter, 4 cm; height, 50 cm). ET production was determined at 12, 24 and 48 h after treatment using the same method as described by Lu et al. 18 . Each plant treatment at each time interval was replicated ten times.

PA content determination.
The experiments were performed twice using two batches of plants. For the first batch of plants, plants were randomly assigned to treatments as follows: WBPH nymph, WBPH gravid female and control. For the second batch of plants, plants were randomly assigned to BPH gravid female and control treatment. Rice sheaths were harvested 3 d after BPH or 4 d after WBPH infestation, and each treatment was replicated 5 times.
Samples (100 mg each) were ground in liquid nitrogen, and PAs were extracted and analyzed using the method as described in Alamgir et al. 15

Direct effects of PAs on WBPH.
Each of the 7 PAs -FerPut, FerTyr, FerAgm, CouAgm, CinPut, DiferSpe and CouPut -was dissolved in methanol (1 mg per ml methanol) and then added to artificial diet at a final concentration of 5 or 100 µg/ml 37 . Fifteen newly emerged WBPH female adults were fed on artificial diet containing one of the 8 PAs at a concentration of 5 or 100 mg L −1 in a 30 ml double-ended open glass cylinder (diameter 2 cm, length 9 cm) as described in Fu et al. 38 . Controls were fed on artificial diet without PAs. Each treatment was replicated seven times. The number of adults alive was recorded every day.
Statistical analysis. Differences in data in different WBPH and BPH treatments were analyzed by Duncan's multiple range test (or Student's t-test for comparing two treatments). All tests were carried out with Statistica 6 (Statistica, SAS Institute Inc., Cary, NC, USA).