An odorant receptor mediates the avoidance of Plutella xylostella against parasitoid

Background Ecosystems are brimming with myriad compounds, including some at very low concentrations that are indispensable for insect survival and reproduction. Screening strategies for identifying active compounds are typically based on bioassay-guided approaches. Results Here, we selected two candidate odorant receptors from a major pest of cruciferous plants—the diamondback moth Plutella xylostella—as targets to screen for active semiochemicals. One of these ORs, PxylOR16, exhibited a specific, sensitive response to heptanal, with both larvae and adult P. xylostella displaying heptanal avoidance behavior. Gene knockout studies based on CRISPR/Cas9 experimentally confirmed that PxylOR16 mediates this avoidance. Intriguingly, rather than being involved in P. xylostella–host plant interaction, we discovered that P. xylostella recognizes heptanal from the cuticular volatiles of the parasitoid wasp Cotesia vestalis, possibly to avoid parasitization. Conclusions Our study thus showcases how the deorphanization of odorant receptors can drive discoveries about their complex functions in mediating insect survival. We also demonstrate that the use of odorant receptors as a screening platform could be efficient in identifying new behavioral regulators for application in pest management. Supplementary Information The online version contains supplementary material available at 10.1186/s12915-024-01862-9.

semiochemicals have become a promising tool for agricultural pest management.At present, there are hundreds of semiochemicals used for population monitoring and pest control.With the continuous development of chemical ecology approaches, research into the interactions between insects and semiochemicals will yield better achievements, serving pest control and production practice.
The semiochemical identification process includes extraction or headspace collection, identification of active compounds, characterization of the chemical composition of the identified compounds, and elucidation of insect behavioral responses to the active semiochemicals.For decades, researchers have relied on conventional chemical ecology approaches based on electrophysiology such as electroantennography (EAG), gas chromatography-electroantennographic detection (GC-EAD), and indoor and/or field behavioral experiments to screen for active semiochemicals to be used in IPM [6,7].Such approaches have been successful in identifying sex pheromone lures for many moths and in developing strong bisexual kairomonal attractants for key crop pests such as the dried fruit beetle, the Mediterranean fruit fly, and the codling moth [8].However, the development of new active semiochemicals requires extensive electrophysiological and behavioral biological assays, which are timeconsuming, sometimes inefficient, and that always rely on the availability of living and healthy insects.Although GC-EAD enables convenient screening of active compounds, a large number of bioassay experiments are still required to avoid false positives [9].For example, a compound can generate an electrical signal but be behaviorally inactive.
Knowledge of the molecular mechanisms underlying insect chemoreception opens innovative routes to identify active odorants via the so-called reverse chemical ecology approach, in which the molecular receptors of semiochemicals are targeted [9,10].These molecular receptors include multiple olfactory proteins, such as odorant-binding proteins (OBPs), chemosensory proteins, odorant receptors (ORs), ionotropic receptors, and sensory neuron membrane proteins [11][12][13].OBPs and ORs have been the focus of previous studies because they are essential for volatile molecule signaling.For example, ORs regulate Drosophila melanogaster recognition of geosmin and pheromone and Helicoverpa armigera recognition of phenylacetaldehyde [14][15][16].OBPs regulate Ailuropoda melanoleuca recognition of pheromones and Culex quinquefasciatus recognition of trimethylamine and nonanal [17,18].These examples show that ORs and OBPs can be used as targets to screen potential semiochemicals for other insects.
The diamondback moth, Plutella xylostella (Lepidoptera: Plutellidae), is a major pest of cruciferous plants and prefers to consume Brassica vegetables.Since 1990, the production of Brassica vegetable in China has increased 20-fold; meanwhile, the area of Brassica vegetable crops damaged by P. xylostella has also increased from 0.15 million ha in 1990 to 2.23 million ha in 2010; this pest is notorious for its strong reproductive ability and high insecticide resistance [19].The annual crop losses and control costs due to P. xylostella represent 4-5 billion USD [20,21].In recent years, the use of semiochemicals has become the primary means for pest management, but there are still relatively few options for the control of P. xylostella because of a lack of candidate odorants [4,19,[21][22][23][24][25].Therefore, we performed an OR-based screen to identify new active odorants for possible application in the control P. xylostella, targeting two peculiar P. xylostella ORs (further referred as PxylORs) expressed in both larval and adult stages.We successfully identified an odorant, heptanal, whose detection is controlled by PxylOR16, and that mediates avoidance behavior of P. xylostella to the parasitoid wasp Cotesia vestalis.Our research not only paves the way for better understanding the mechanisms of the avoidance behavior to parasitoid wasps in this species but also highlights the efficient use of an OR to identify a behaviorally active volatile, providing an innovative approach for OR-based screening of potential behavioral regulators (repellents and attractants) for application in pest management.

PxylOR16 is expressed in all the larval and adult stages of P. xylostella and specifically responds to heptanal
Most insect ORs are expressed in the adult antennae and, usually, only a few ORs are expressed in larvae.Some ORs expressed in both larvae and adults have been shown to exert key functions, for instance in D. melanogaster, Spodoptera littoralis, and H. armigera [16,26,27].We thus searched for such ORs in P. xylostella, as targets for large screening to identify new active volatile compounds.First, the expression profile of 54 PxylORs in the heads (including antennae) of larvae at different instars (first, second, third female, third male, fourth female, and fourth male) and in the antennae of female and male adults were determined (Fig. 1A, Additional File 1: Fig. S1).The P. xylostella ortholog of the universal OR co-receptor (Orco), which is necessary for OR functioning, was found to be expressed in all samples examined.Most pheromone receptors (PRs) were specifically or highly expressed in the male antennae (PxylOR1, Pxy-lOR3, PxylOR4, PxylOR5, PxylOR6, PxylOR7, and Pxy-lOR41) (Additional File 1: Fig. S1).Interestingly, only two ORs, PxylOR16 and Pxy-lOR27, were expressed in all examined larval stages and adult antennae of both sexes (Fig. 1A), suggesting they may have important biological functions throughout the insect life cycle.We subsequently expressed these two ORs in Xenopus oocytes to investigate their responsivity to a panel of 71 plant volatile compounds (Additional File 1: Table S1) using two-electrode voltage clamp.The oocytes expressing PxylOR16/Orco were extremely sensitive to heptanal, and they also responded weakly to two other volatiles (octyl acetate and 1-heptanol) (Fig. 1BE).In dose-response studies, we assayed the responses of PxylOR16 to a range of concentrations of heptanal and observed the lowest measurable response at a concentration of 1 × 10 −6 M (Fig. 1D).The EC 50 value of PxylOR16 for heptanal was 1.757 × 10 −5 M (Fig. 1F).These results suggest that Pxy-lOR16 and its ligand (heptanal) are of great significance to P. xylostella.Oocytes expressing PxylOR27 did not respond to any of the 71 examined plant volatile compounds (Fig. 1C).

PxylOR16 knockout mutants show impaired electrophysiological responses to heptanal
Since heptanal appeared to strongly activate PxylOR16, we wondered whether this compound would also trigger a strong antennal response in vivo, as measured by EAG.Because the antennae of P. xylostella larvae are too small, we were not able to carry out EAG experiments on larvae.We thus recorded the electrophysiological responses of adult female and male antennae to heptanal doses of 10 ng, 100 ng, 1 μg, and 10 μg (Fig. 2B, left).Both female and male antennae displayed a dose-dependent EAG response to heptanal, with responses increasing with increasing heptanal doses.At the highest dose (10 μg), the EAG response values were maximal, with mean response values of 0.77 mV and 0.52 mV for wild-type females and males, respectively.At all doses except the lowest one (10 ng), the male EAG response values were significantly higher than those of females (Fig. 2B, left).We next constructed a PxylOR16-knockout strain using CRISPR/Cas9 genome editing in a traditional and common widely used method in many insects including P. ).Left, dose-dependent electroantennographic (EAG) responses of female and male moths.Heptanal was used at doses ranging from 10 ng to 10 μg.WT female and male antennae exhibited dose-dependent EAG responses to heptanal, with responses increasing with increasing heptanal doses.EAG responses of female and male PxylOR16 −/− moths to heptanal were far lower than those of WT female and male moths at all heptanal doses.Right, as a control, we tested the responses of female and male moths to trans-2-hexen-1-ol, which is not a ligand of PxylOR16.As expected, we observed no differences in the responses to this compound between PxylOR16 −/− moths and WT moths.Error bars indicate SEM (n = 10).Different letters indicate significant differences among insects (two-way ANOVA followed by Tukey's pairwise test; P < 0.05) xylostella [28], H. armigera [16], Locusta migratoria [24,29], and Eupeodes corollae [30].In all of these works, only one homozygous mutant strain was used.Therefore, knockout mutants (PxylOR16 −/− ) were obtained with a 5-nt insertion and a 1-nt deletion in the second exon, which introduced a premature stop codon in the coding sequence (Fig. 2A).This PxylOR16-knockout P. xylostella strain (PxylOR16 −/− ) was using for the further comparing of electrophysiological and behavioral responses with the wild-type (WT) strain.We also used EAG to investigate the electrophysiological response of PxylOR16 −/− moths to heptanal.As a control, we tested the response of female and male mutant moths to trans-2-hexen-1-ol, which is not a ligand of PxylOR16, and that is known evoke significant electrophysiological responses in the antennae of both female and male P. xylostella adults.As expected, we observed no difference between the responses of PxylOR16 −/− moths and WT moths to this compound (Fig. 2B, right).However, the EAG signals of female and male PxylOR16 −/− moths in response to heptanal were significantly reduced compared with those of WT female and male moths, at all examined heptanal doses (Fig. 2B, left).In the abovementioned results, the EAG responses to heptanal and trans-2-hexen-1-ol at different concentrations have been registered on the same individual, whether WT or PxylOR16 −/− .These results support the hypothesis that PxylOR16 is involved in heptanal detection.

Heptanal elicits obvious avoidance behaviors in P. xylostella larvae and adults
As PxylOR16 was expressed in both larvae and adults, and having confirmed that PxylOR16 responded to heptanal, we next investigated the effect of this ligand on the behavior of P. xylostella adults and larvae.Behavioral experiments were conducted independently on female and male adults.Adult behavioral responses were tested using a Y-tube two-choice bioassay, while larvae responses were tested using a Petri dish assay (static air).As a control, we selected sex pheromone (Z-11-hexadecenal) to verify whether there is a difference in recognition of sex pheromones between the WT and PxylOR16 −/− .The results of behavioral experiments showed that both WT and PxylOR16 −/− male moths were significantly attracted to Z-11-hexadecenal, and there was no significant difference between the two strains (Additional File 1: Fig. S2).
According to the Y-tube two-choice bioassay for adults, we first validated this assay by performing control experiment in which insects were exposed to filter paper without any odorant added on both sides.We observed that neither WT nor PxylOR16 −/− adults displayed any selection preference (Fig. 3A).Then, we tested adult behavioral responses to heptanal at different concentrations.The experiment revealed that male moths displayed significant avoidance to heptanal compared with the paraffin oil control at doses of 100 ng, 1 μg, and 10 μg (Fig. 3A, left).The same avoidance behavior was observed for female moths, but only at doses of 1 μg and 10 μg (Fig. 3A, right), suggesting that female and male moths have different sensitivities to heptanal, in accordance with the EAG results.
Then, we performed a second experiment on adults using the Y-tube two-choice bioassay to test the response of P. xylostella adults to heptanal in the presence of the host Brassica pekinensis leaf (1 × 1 cm 2 , the oviposition material for adults used in our laboratory) as the attractant in both tube arms.In the one side blank experiment, we added a piece of B. pekinensis in one arm and kept the other arm blank.Both WT and PxylOR16 −/− adults exhibited a preference for B. pekinensis over the blank (Fig. 3B, Blank).This result showed that the B. pekinensis leaf was attractive to adults, whatever the genotype.When adults were challenged with B. pekinensis on both sides, neither WT nor PxylOR16 −/− adults displayed any selection preference (Fig. 3B, Control).When we compared the choice between B. pekinensis and B. pekinensis + heptanal, we found that WT male moths were significantly less attracted to B. pekinensis + heptanal than to B. pekinensis alone (with paraffin oil), at heptanal doses of 100 ng, 1 μg, and 10 μg (Fig. 3B, left).Female WT moths were also less attracted to B. pekinensis + heptanal than to B. pekinensis alone, but only at heptanal doses of 1 μg and 10 μg (Fig. 3B, right), indicating that male moths were still more sensitive to heptanal than female moths in the presence of B. pekinensis.Consistent with our finding that PxylOR16 mediates P. xylostella perception of heptanal, PxylOR16 −/− adults in the Y-tube two-choice bioassay showed no preference under any examined heptanal concentration (Fig. 3AB).
Larvae behavioral responses were tested using a Petri dish assay (static air).As for the Y-tube assay conducted in adults, we first conducted a control one side blank experiment in which a piece of B. pekinensis (1 × 1 cm 2 , the food source for larvae used in our laboratory) was added on one side of the dish, while keeping the other side blank.This result showed the food source was also attractive to larvae (Fig. 3C, Blank).When larvae were challenged with B. pekinensis on both sides, neither WT nor PxylOR16 −/− larvae displayed any selection preference (Fig. 3C, Control).In view of these results, we used the Petri dish assay to test the response of P. xylostella larvae to heptanal (heptanal alone vs. paraffin oil control) in the absence of food (B.pekinensis) and found that P. xylostella larvae did not make effective choices.We next tested the response of P. xylostella larvae to heptanal in the presence of a piece of B. pekinensis using the Petri dish assay.Significantly more female and male third instar larvae were located in the control area (B.pekinensis + paraffin oil) than in the heptanal-supplemented area (B.pekinensis + heptanal) at all heptanal doses (100 ng, 1 μg, and 10 μg on filter paper) except at 10 ng (Fig. 3C).To verify whether PxylOR16 was responsible for this effect in the larvae, we challenged PxylOR16 −/− larvae in Fig. 3 Heptanal elicits obvious avoidance behaviors in both Plutella xylostella larvae and adults.Preference index of P. xylostella wild-type (WT) and CRISPR/Cas9 PxylOR16 knockout (PxylOR16 −/− ) adults and larvae for heptanal.Preference indexes = (number of choices at treatment -number of choices at control) / number of total choices.A Preference index of WT and PxylOR16 −/− mutant female and male moths for control (filter paper vs filter paper) and for filter paper + paraffin oil vs filter paper + heptanal in a Y-tube olfactometer (n = 16).WT male moths exhibited significant avoidance to heptanal at doses of 100 ng, 1 μg, and 10 μg (left).Avoidance was also observed for WT female moths, but only at doses of 1 μg and 10 μg (Welch's t-test; NS, no significant difference, P > 0.05; *** P < 0.001) (right).PxylOR16 −/− mutant female and male moths did not show a preference for one side of the device (Welch's t-test; NS, no significant difference, P > 0.05).B Preference index of WT and PxylOR16 −/− mutant female and male moths for blank (one side blank vs a piece of Brassica pekinensis), for control (a piece of B. pekinensis vs a piece of B. pekinensis), and for a piece of B. pekinensis + paraffin oil vs a piece of B. pekinensis + heptanal in a Y-tube olfactometer (n = 16).WT and PxylOR16 −/− mutant female and male moths were significantly attracted to B. pekinensis, compared with blank.WT male moths were significantly less attracted to B. pekinensis + heptanal at doses of 100 ng, 1 μg, and 10 μg than to B. pekinensis alone (with paraffin oil) (left).Female moths were also less attracted to B. pekinensis + heptanal than to B. pekinensis alone but only at heptanal doses of 1 μg and 10 μg (Welch's t-test; NS, no significant difference; P > 0.05; *** P < 0.001) (right).The PxylOR16 −/− mutant female and male moths lost their avoidance responses to heptanal significantly, compared with WT adult female and male moths (Welch's t-test; NS, no significant difference; P > 0.05).C Preference indexes of 10 WT and mutant female and male third instar larvae for blank (one side blank vs a piece of B. pekinensis), for control (a piece of B. pekinensis vs a piece of B. pekinensis), and for a piece of B. pekinensis + paraffin oil vs a piece of B. pekinensis + heptanal in a 10-cm diameter plastic Petri dish (n = 28).Both sexes of WT and mutant larvae were attracted to B. pekinensis.Significantly more female and male third instar larvae were located in the control area (B.pekinensis + paraffin oil) than in the heptanal-supplemented area (B.pekinensis + heptanal) at all heptanal doses (100 ng, 1 μg, and 10 μg on filter paper) except for 10 ng (Welch's t-test; NS, no significant difference, P > 0.05; *** P < 0.001).As shown for PxylOR16 −/− adults, detection of heptanal was abolished in PxylOR16 −/− larvae (Welch's t-test; NS, no significant difference, P > 0.05) the Petri dish assay.As shown for PxylOR16 −/− adults, detection of heptanal was abolished in PxylOR16 −/− larvae (Fig. 3C).Taken together, these results suggest that PxylOR16 regulates the heptanal avoidance behavior of P. xylostella larvae and adults.

Heptanal does not participate in the direct interaction between P. xylostella and its host plant
Heptanal is a common plant volatile, so we firstly speculated that heptanal may mediate interactions between P. xylostella and host plants.We conducted a series of behavioral experiments to test if heptanal participates in the direct interaction between P. xylostella and Brassica vegetables.We chose Brassica plants for experiments because it is the most important genus of cruciferous plants and also the favorite host plant of P. xylostella.In the past 20 years, P. xylostella has become the most destructive insect pest on Brassica vegetables [21].Therefore, we selected a common Brassica vegetable (B.parachinensis) for the following experiments.
The effect of heptanal in the presence of the host plant B. parachinensis on the adults and larvae were tested by using Y-tube two-choice bioassay and a screening device (with airflow) separately.The responses of P. xylostella adults and larvae to a healthy B. parachinensis plant were tested in one side blank experiments.The healthy B. parachinensis plant did have a significant attracting effect on both WT and PxylOR16 −/− adults and larvae, and neither WT nor PxylOR16 −/− insects displayed any preference in control experiments when B. parachinensis was present in both sides (Fig. 4AB).
In subsequent experiments, female and male moths were given a choice between an intact B. parachinensis plant without and with different doses of heptanal in a Y-tube assay.Three doses (100 ng, 1 μg, and 10 μg) and two doses (1 μg and 10 μg) of heptanal elicited avoidance behaviors in female and male P. xylostella adults, respectively (Fig. 4A).In larvae, we observed a preference of female and male larvae (third instar larvae) for a healthy B. parachinensis plant with paraffin oil over a healthy plant with heptanal at three doses (100 ng, 1 μg, and 10 μg) (Fig. 4B).The numbers of PxylOR16 −/− adults and larvae on each side of the device were not significantly different at all heptanal doses tested (Fig. 4AB).To test whether heptanal is present in the host plant B. parachinensis, we examined the volatiles emitted by the healthy B. parachinensis.Briefly, we used solid-phase microextraction to collect volatiles emitted by healthy plants.Gas chromatography coupled to mass spectrometry (GC/ MS) analysis revealed that no heptanal was detectable in healthy B. parachinensis volatiles (Fig. 4C).
We found that an intact B. parachinensis plant had a significant attracting effect on P. xylostella and that its presence did not affect the heptanal avoidance behavior, so we further tested if the B. parachinensis plants damaged by P. xylostella feeding will release heptanal and evoke avoidance behavior.Female and male moths and third instar larvae were given a choice between an intact B. parachinensis plant and larval-infected B. parachinensis plant (larvae and feces were removed, same treatment in following experiments) in a Y-tube and in a screening device: the adults and larvae had a preference for larvalinfected plants compared with heathy plants (Fig. 4DE).These results were not unexpected, since P. xylostella was previously reported to prefer P. xylostella larvalinfected plants [31,32].Moreover, like the WT, the Pxy-lOR16 −/− adults and larvae also showed a preference for larval-infected plants (Fig. 4DE).Next, we also examined the volatiles emitted by the P. xylostella larval-infested B. parachinensis.GC/MS analysis revealed that no heptanal was detectable in larval-infested B. parachinensis volatiles (Fig. 4C).
These results suggest that heptanal cannot explain the direct interaction between P. xylostella and B. parachinensis (intact or pest infested).Similarly, many other Brassica plants do not release heptanal [33], so we speculate that heptanal is not involved in the interactions between P. xylostella and host plants as a plant volatile.

P. xylostella larvae and adults detect and avoid Cotesia vestalis odorants
Heptanal was previously reported to be a cuticular volatile emitted and sensed by some Cotesia parasitic wasp species, including C. glomerata and C. marginiventris, and this compound is a required component of their sex communication system [34,35].Both females and males of C. glomerata, a parasitic wasp of Pieris brassicae, release heptanal as a repellent pheromone.After mating, females release heptanal to repel males to avoid mating again, while the males release heptanal to repel conspecific males, thus reducing male-male competition on the natal patch and ensuring maximal mating success.The amount of heptanal released by C. glomerata was analyzed.It was previously found that about 3 ng can be obtained from 6 individuals by solvent extraction, and that about 50 ng can be obtained from 200 individuals by headspace extraction [34].In a phylogeny of 25 species of Cotesia commonly used in laboratory and field research, C. vestalis is the closest to C. glomerata [36].We therefore turned our attention to C. vestalis, which is an endoparasitoid wasp that is known to attack P. xylostella larvae; indeed, it is used as an agricultural control measure against P. xylostella [37].
We hypothesized that, like C. glomerata, C. vestalis releases heptanal.To test this hypothesis, we collected cuticular compounds of this wasp using dichloromethane for GC/MS analysis.Consistent with the hypothesis that C. vestalis releases heptanal, we could detect heptanal in the extracts of a mixture of 10 female and 10 male wasps, and the amount collected from these twenty wasps was 2.47 ± 0.55 ng (Fig. 5AB).
Both D. melanogaster adults and larvae have an avoidance response to Leptopilina boulardi, which is a parasitic wasp of D. melanogaster larvae [26].To test whether larvae and adults of P. xylostella have a similar avoidance response to C. vestalis odorants, we next ).WT and PxylOR16 −/− mutant female and male moths were significantly attracted to healthy B. parachinensis plant, compared with blank.Three doses (100 ng, 1 μg, and 10 μg) and two doses (1 μg and 10 μg) of heptanal elicited avoidance behavior in WT female and male P. xylostella adults, (Welch's t-test; NS, no significant difference, P > 0.05; *** P < 0.001).The numbers of PxylOR16 −/− adults on each side of the Y-tube olfactometer were not significantly different at all heptanal doses tested, when comparing B. parachinensis + heptanal with B. parachinensis plant + paraffin oil (Welch's t-test; NS, no significant difference, P > 0.05).B Preference indexes of 10 WT and mutant female and male third instar larvae for blank (one side blank vs an intact healthy B. parachinensis plant), for control (an intact healthy B. parachinensis plant vs an intact healthy B. parachinensis plant), and for an intact healthy B. parachinensis plant + paraffin oil vs an intact healthy B. parachinensis plant + heptanal in a screening device (n = 10).Both sexes of WT and mutant larvae were attracted to the B. parachinensis plant.WT female and male larvae (third instar larvae) preferred the healthy plant with paraffin oil to a healthy plant with heptanal at three doses (100 ng, 1 μg, and 10 μg), unlike PxylOR16 −/− larvae (Welch's t-test; NS, no significant difference, P > 0.05; *** P < 0.

001). C GC-MS analysis of the volatiles produced by larval-infested plants and healthy plants; heptanal was not detected. Larval-infested B. parachinensis (top), healthy B. parachinensis (middle), and synthetic heptanal (bottom). D Preference index of WT and PxylOR16 −/− mutant adults for control (intact healthy B. parachinensis plant vs intact healthy B. parachinensis plant) and for an intact healthy B. parachinensis plant vs a larval-infected B. parachinensis plant in a Y-tube olfactometer (n = 16
).There was no difference in selection between the WT and PxylOR16 −/− mutant; both preferred the larval-infected plant (Welch's t-test; NS, no significant difference, P > 0.05; *** P < 0.001).E Preference index of WT and PxylOR16 −/− mutant larvae for control (intact healthy B. parachinensis plant vs intact healthy B. parachinensis plant) and for an intact healthy B. parachinensis plant vs a larval-infected B. parachinensis plant in a screening device (n = 10).Like adults, WT and PxylOR16 −/− mutant larvae both preferred the larval-infected plant (Welch's t-test; NS, no significant difference, P > 0.05; *** P < 0.001) conducted a behavioral experiment in two different experimental treatments: (1) a C. vestalis body wash obtained by extracting 20 wasps (females and males) with dichloromethane for 30 min and (2) 20 live C. vestalis wasps (females and males) placed in glass jars connected by a Y-tube or a screening device.For the first experimental treatment, a Y-tube two-choice bioassay and a Petri dish assay revealed that a C. vestalis body wash elicited avoidance responses in adults (Fig. 5C, left) and in larvae (Fig. 5C, right).Both adults (Fig. 5D, left) and larvae (Fig. 5D, right) of P. xylostella also showed avoidance behavior in response to live C. vestalis in a Y-tube two-choice bioassay and using a screening device.
In behavioral experiments using PxylOR16 −/− , we found that both PxylOR16 −/− larvae and adults no longer had an avoidance response to C. vestalis body wash or live C. vestalis (Fig. 5CD).These results show that P. xylostella larvae and adults use PxylOR16 to detect heptanal released by C. vestalis and thereby to regulate their avoidance to C. vestalis.

Discussion
In this study, we identified two P. xylostella OR expressed at all developmental stages.One of these ORs, PxylOR27, did not respond to any of the 71 examined plant volatile compounds.We thus speculate that the ligand(s) of Pxy-lOR27 do not belong to the compound panel we tested here or that PxylOR27 did not successfully express in oocyte membrane [16,38].The other OR, PxylOR16 expressed at all developmental stages and detected under the sensilla trichoidea of male and female antennae (Additional File 1: Fig. S3), was activated by heptanal.This receptor appeared to be narrowly tuned, a characteristic of ORs tuned to ecologically relevant stimuli [15,38,39] and involved in labeled line olfactory circuits [40].We thus suspected that heptanal may play an important role in P. xylostella and searched for its function.
We investigated whether heptanal could be an indicator of an unsuitable environment for P. xylostella.Heptanal is known to have different functions in multiple species.It has been identified as a component of linden (Tilia tuan), which can elicit electroantennographic and behavioral responses in Agrotis ipsilon.Heptanal is an important compound released by peach flowers and mature fruits, and it also elicits electroantennographic responses in Grapholita molesta and attracts this species [41][42][43].Similarly, heptanal is a novel identified pheromone, which is derived from the cuticular hydrocarbons of D. melanogaster and plays an important role in the evaluation of oviposition site [44].Heptanal can also have different bioactivities depending on its concentration.For example, in Phthorimaea operculella, it acts as an oviposition repellent at high concentrations and as an attractant at low concentrations [45].Interestingly, most cruciferous plants-the host plants of P. xylostella-do not release heptanal [33].We found that neither healthy nor larval-infested Brassica could repel P. xylostella.On the contrary, P. xylostella larvae and adults preferred larval-infested plants over healthy plants, which is consistent with previous studies [31,32].This is because the release of isothiocyanates increases after Brassica are damaged, and isothiocyanates are highly attractive to P. xylostella [46,47].Thus, the ecological significance of the avoidance behavior elicited by heptanal does not lie in the interaction between P. xylostella and Brassica, but may be related to the detection of other dangers, such as the presence of natural enemies.
We then searched for other sources of heptanal in the P. xylostella environment.We focused on natural enemies such as endoparasitoid wasps, since heptanal is known to be emitted by some Cotesia species.Heptanal has been shown to function directly in the pheromonal communication of C. glomerata and C. marginiventris [34,35].Our experiments showed that larvae and adults of P. xylostella avoid heptanal.This shows that among P. brassicae-C.glomerata-P.xylostella in the same ecosystem, heptanal belongs to a "semiochemical parsimony", which refers to the use of a chemical substance for two or more purposes [48].Another reason for the release of heptanal by C. glomerata may be to protect P. brassicae, leading to a better feeding environment.A similar situation also occurs in the interaction among D. melanogaster, Leptopilina heterotoma, and ants.L. heterotoma is a parasitic wasp of D. melanogaster and releases iridomyrmecins to repel ants; this defense mechanism allows the parasitic wasp to forage without interference from the ants, thereby improving their parasitism and reproductive success.At the same time, iridomyrmecins also take part in the mating behavior of L. heterotoma [49,50].
For instance, C. marginiventris females can release heptanal to significantly improve their attractiveness to males [35].C. vestalis is a Cotesia species known to be an endoparasitoid of P. xylostella.Heptanal is an attractant for C. vestalis, and this attractant has been used successfully in the biological control of P. xylostella [37].We thus hypothesized that C. vestalis may also release heptanal, and we confirmed this hypothesis by revealing that this compound was a component of a C. vestalis cuticular extract.Taken together our results, we propose that heptanal serves as a signal for both P. xylostella adults and larvae to recognize natural enemies such as C. vestalis and avoid parasitization.Further experiments, such as testing the effect of heptanal and the role of PxylOR16 on the parasitism rate of P. xylostella in the presence of C. vestalis, would definitely confirm this hypothesis.It has to be noticed that P. xylostella avoidance behavior was observed at high wasp density (20 wasps were needed to elicit avoidance).One explanation could be that a trade-off exists between the need to feed/oviposit and parasitization risk: at low wasp concentration, feeding or oviposition would be prioritized, whereas at high was concentration, escape is prioritized.
Nevertheless, our results suggest an essential purpose of heptanal in Cotesia parasitoids as it is released by at least three different Cotesia species.There are more than 350,000 species of parasitoid wasps on earth [51], indicating that they are an indispensable part of the self-regulation of the ecosystem.Insects have evolved a series of defenses, some cellular and some behavioral, to protect themselves against these parasitoids [52,53].D. melanogaster regulates defensive behaviors against parasitic wasps through different kinds of neurons.For example, nociceptive neurons mediate rolling behavior [49], visual cues regulate oviposition behavior [54,55], and olfactory neurons detect odors of parasitic wasps and drive aversion responses [26].Studies have shown that the presence of certain parasitoid wasps can affect the mating behavior of D. melanogaster [56].In addition, the insect olfactory system plays an important role in the detection of parasitic wasps, and there are many examples of species using allomones produced by natural enemies to escape or to choose habitats without, or that are not favorable to, natural enemies to feed and lay eggs [26,57,58].Therefore, accurately determining the presence or absence of natural enemies is crucial to the life activities of insects.P. xylostella may use heptanal as a primary indicator for judging the presence of C. vestalis or C. glomerata in the same ecological environment.The use by a species of odors released by another species for sexual communication to avoid natural enemies is highly adaptive because the odor-based sex communication system cannot be easily changed [26].The phenomenon of insects using odorants of enemies for avoidance is in line with the strategy used by P. xylostella against its enemy.
Our study provides a reference for OR-based screening of novel behavioral regulators of P. xylostella, which could serve as a scientific basis for formulating and improving integrated pest control strategies.In addition, we gained a deeper understanding of how insect pests recognize volatiles produced by their natural enemies through olfaction.Better knowledge of the relationship between the host plants, parasitoid wasps, heptanal, and P. xylostella is now needed to understand the balance between preference and repellency, especially in females.Such knowledge would be useful in the control of P. xylostella infestation by enhancing indirect plant defense via modification of the heptanal synthetic pathway or by using adequate amounts of synthetic heptanal to repel P. xylostella.

Conclusions
We identified heptanal as a repellent for both larvae and adults of P. xylostella, via the functional characterization of the odorant receptor PxylOR16.Our findings shed new insights on the interaction mechanism among insect pests and their natural enemies and provides a candidate compound that could be used for integrated pest management of this important agricultural pest.

Plants and insect rearing
P. xylostella larvae were obtained from a colony maintained at the Institute of Plant Protection, Chinese Academy of Agricultural Sciences, which was collected from a suburban area in Beijing in 2001.The larvae were reared on B. pekinensis, and the adults were fed with 10% honey water at 26 ± 2 ℃ on a 16:8-h (light/dark) photoperiod cycle and 50-70% relative humidity (RH).C. vestalis cocoons were obtained from the Institute of Insect Sciences, Zhejiang University, China.Adult wasps were propagated using third instar P. xylostella larvae, and the rearing of C. vestalis colonies was conducted following previously reported protocols [59].Seedlings of B. parachinensis Bailey were cultivated in a growth chamber (25 ± 2 ℃, 50-70% RH, 12:12 light/dark regime) for 5-6 weeks up to the six-leaf stage.

Expression analyses
Larval heads (first, second, third female, third male, fourth female, and fourth male instars) and female and male adult antennae of P. xylostella were collected.It is easy to distinguish the sexes of third and fourth instar larvae, as male larvae have testes in the fifth abdominal segment, and their backs are pale yellow.Total RNAs were extracted separately from each collected tissue using TRIzol Reagent (Invitrogen, Carlsbad, CA) and treated with DNaseI to remove trace amounts of genomic DNA.First-strand cDNAs were synthesized using the RevertAid First Strand cDNA Synthesis Kit (Fermentas, Glen Burnie, MD, USA) and were used as templates RT-PCR.The expression patterns of all PxylORs were investigated.The primer sequences are listed in Additional File 1: Table S2.PCR conditions: 94 ℃ for 3 min, then 35 (larvae) or 26 (adults) cycles at 94 ℃ for 30 s, 55 ℃ for 40 s, and 72℃ for 60 s, with a final 10-min incubation at 72 ℃.PCR products were analyzed on 2.0% agarose gels.The experiments were repeated three times with independent RNA samples.

Cloning of PxylOR16 and PxylOR27
The open reading frames of PxylOR16 and PxylOR27 were cloned using specific primers (Additional File 1: Table S2).PCR amplification products were purified from 1% agarose gels and ligated into the pEASY-Blunt vector (TransGenBiotech, China).After transformation of Trans1-T1 competent cells (TransGenBiotech, China), selected positive clones were sequenced by BGI (Beijing, China).

Plant volatile compounds
Seventy-one odorants, including host-plant volatiles and known behaviorally and physiologically active semiochemicals, were used for functional characterization of PxylORs.All compounds were purchased from Sigma-Aldrich (purity ≥ 98%) and are listed in Additional File 1: Table S1.For functional analyses of PxylORs using two electrode voltage clamp electrophysiological recording, compounds were diluted to 1 M with dimethyl sulfoxide (DMSO) as stock solutions and stored in − 20 ℃ until use.The stock solutions were diluted to 10 -4 M with 1 × Ringer's buffer for ligand screening of PxylORs.

Receptor expression in Xenopus oocytes and two electrode voltage clamp electrophysiological recording
Xenopus oocyte collection and preparation and electrophysiological recordings were conducted following previously reported protocols [60,61].The PxylOR16, PxylOR27, and PxylOrco cRNAs were synthesized using mMESSAGE mMACHINE T7 (Ambion, Austin, TX).PxylOR16 or PxylOR27 cRNAs (27.6 ng each) were simultaneously microinjected with PxylOrco cRNA into mature healthy Xenopus oocytes (stages V-VII).After injection, oocytes were incubated for 4-7 days at 16 ℃ in 1 × Ringer's solution supplemented with 5% dialyzed horse serum, 50 μg/mL tetracycline, 100 μg/mL streptomycin, and 550 μg/mL sodium pyruvate.Odorant-induced currents were recorded with an OC-725C oocyte clamp (Warner Instruments, Hamden, CT) at a holding potential of − 80 mV.Oocytes were exposed to compounds in ascending order of concentration with an interval between exposures that allowed the current to return to baseline.The data were acquired and analyzed with Digidata 1440A and pCLAMP 10.2 software (Axon Instruments Inc., Union City, CA).Statistical comparison of responses of oocytes to tested ligands and doseresponse data were analyzed using GraphPad Prism 5 (GraphPad Software Inc., San Diego, CA).

Homozygous mutant generation by CRISPR/Cas9
To generate PxylOR16-knockout mutants, we selected a small sequence (GGC GCC CGA GCT GCA AGC TAT CAA GG) as the single-guide RNA (sgRNA) target site in the second exon of PxylOR16.We have analyzed the possible off-target effects of this sgRNA in the P. xylostella genome and found that the sgRNA targets a single site, indicating that this guide has high specificity.sgRNA was synthesized according to the manufacturer's procedures (GeneArt ™ Precision gRNA Synthesis Kit, ThermoFisher Scientific, Pittsburgh, PA, USA).For sgRNA synthesis, the upstream primer (TAA TAC GAC TCA CTA TAG + target sequence) and the downstream primer (TTC TAG CTC TAA AAC + target sequence reverse complement) were spliced by PCR with a mixture of the Tracr fragment and the T7 primer to obtain the DNA template of the sgRNA.Then the sgRNA was synthesized by in vitro transcription.The PCR assembly reaction and in vitro transcription were conducted according to the manufacturer's instructions.Finally, the DNA template was digested with DNase I and the sgRNA was cleaned using the sgRNA Clean Kit (ThermoFisher Scientific, Pittsburgh, PA, USA).The primer sequences are listed in Additional File 1: Table S2.
Based on the PxylOR16 genomic sequence, upstream (PxylOR16F) and downstream (PxylOR16R) primers were designed to detect mutants using PCR with genomic DNA as a template (Additional File 1: Table S2).Genomic DNA was extracted from adults using a TIANamp Genomic DNA Kit (Tiangen, China).PCR amplification products were purified on a 1.5% agarose gel, ligated into the pEASY-Blunt vector (TransGenBiotech, China), and sequenced by BGI (Beijing, China).
Freshly laid eggs (within 30 min) were used for microinjection.About 1 nL of a mixture of Cas9 nuclease (600 ng/µL, GeneArt Platinum Cas9 Nuclease, Ther-moFisher Scientific, Pittsburgh, PA, USA) and sgRNA (600 ng/µL) was microinjected into each egg using a FemtoJet and InjectMan NI 2 microinjection system (Eppendorf, Hamburg, Germany).Approximately 500 eggs were microinjected, and 80 larvae hatched (G0).The adults of the G0 generation were hybridized to generate G1 individuals.When these G1 individuals grew into adults, we randomly grouped 50 pairs of parents and each pair was put in a plastic cup for mating and egg laying.As we should keep the insects alive for oviposition, genomic DNA was extracted from the whole bodies of the parents after oviposition for further genotyping by PCR and sequencing.One pair of G1 parents was selected because both the males and females exhibited heterozygous frame-shifting mutations (Fig. 2A), and their offspring was kept as G2.When the G2 individuals grew into adults, 10 pairs were randomly selected to perform single-pair mating and then genotyping by PCR and sequencing.The offspring of one pair of G2 parents of which both male and female carried the same heterozygous mutations were selected and kept as G3.When these G3 individuals grew into adults, 15 pairs were randomly selected to perform single-pair mating and then genotyping by PCR and sequencing.The offspring of one pair of G3 parents of which both male and female carried the same homozygous mutation were selected and kept as G4.The offspring of the G4 homozygous mutants were used for electrophysiological and behavioral analyses (Additional File 1: Fig. S4).

EAG recordings
The responses of the antennae of virgin female and male moths (1-2 days after eclosion) to heptanal and trans-2-hexen-1-ol were analyzed by EAG recording.Heptanal and trans-2-hexen-1-ol were diluted in hexane at 1, 10, 100, and 1000 ng/μL.Test solutions (10 μL each) were deposited on a filter paper strip (1 × 5 cm) that was inserted into a Pasteur pipette.Hexane was used as a control.Each experiment was repeated 10 times.EAG responses for each compound were calculated by subtracting the EAG response to hexane.Data are reported as mean ± standard error of the mean (SEM), and means were compared by two-way ANOVA with Tukey's pairwise test.

Behavioral response trial
The behavioral responses of unmated P. xylostella adults to heptanal, body washes of parasitoid wasps, and live parasitoid wasps were tested using a Y-tube assay and those of larvae were tested using a Petri dish assay and a screening device.Heptanal was dissolved in paraffin oil at different concentrations (1, 10, 100, and 1000 ng/μL), and all bioassays were conducted in a temperature-controlled room at 26 ± 2 ℃.The adult and larvae preference indexes (PI) were calculated as described in previous studies in mosquitoes and Drosophila [62,63] using the following equation: (T − C)/(T + C), where T is the number of choices for the treatment site and C is the number of choices for the control site.
The Y-tube consisted of a 15-cm stem and two 24.5cm arms at a 45° angle with an interior diameter of 2.0 cm.The rate of airflow was 0.5 L/min and was controlled by connecting the olfactory arms to a vacuum pump.Before entering the tube, humidified continuous air was filtered through activated charcoal.The Y-tube was installed inside a behavior box (100 × 60 × 80 cm) equipped with a fluorescent lamp to provide light.A filter paper strip containing 10 μL of test solution was placed in the "treatment" arm, and a filter paper strip containing 10 μL of solvent was placed in the other arm as a control.Six moths were put one by one into the stem of the Y-tube for 10 min.During this time, moths entering one of the two arms more than 5 cm and staying there for more than 30 s were considered have made a choice.Moths that did not make a choice were also recorded but the number was not considered in the analysis.Experiments were repeated 16 times for females and males separately.In Y-tube experiments with B. pekinensis leaves (1 × 1 cm) or intact B. parachinensis plants, leaves and plants were added in both arms and spiked with heptanal or paraffin oil as described above.The preference indexes of female and male moths were compared using Welch's t-test analysis.
We used a simple Petri dish assay and a screening device to test the effect of heptanal, parasitoid wasp body washes, and live parasitoid wasps on P. xylostella third instar larvae.In the Petri dish assay (10 cm diameter × 1.5 cm height), two 1 × 1 cm pieces of fresh B. pekinensis leaves were placed at two opposite points of the Petri dish; one piece was spiked with 10 μL of test solution and the other with 10 μL of solvent alone as a control.After 4 h of starvation, 10 third instar larvae (all males or all females) were placed in one side of the Petri dish.The number of larvae on each piece of fresh B. pekinensis leaves at the opposite sides of the Petri dish was counted after 30 min [64].The screening device (21 × 21 × 4 cm) consisted of three parts: traps A and B, a selection hole, and odor selection area.Traps A and B were each connected with B. parachinensis plants or live parasitoid wasps in glass jars, and traps captured the test larvae.The selection hole linked the trap and odor selection area, and there was a height difference between the trap and odor selection area, which prevented larvae from returning to the odor selection area.The odor selection area had a smaller height, which was conductive to the diffusion of odorant molecules, and a larger area, which provided sufficient activity space and improved the selection accuracy.The rate of airflow was 0.3 L/min and was controlled by connecting the screening device to a vacuum pump.Before entering the glass jars, humidified continuous air was filtered through activated charcoal.The release of test solution was performed in the same way as described for the adult Y-tube experiment (a filter paper strip containing 10 μL of test solution placed in the "treatment" jar and a filter paper strip containing 10 μL of solvent placed in the other jar as a control), with the presence a piece of B. pekinensis or an intact entire plant.Ten starved third instar larvae (all males or all females) were placed in the odor selection area of the screening device.The number of larvae in different sides of the screening device was counted after 2 h.Experiments were repeated 28 times (Petri dish assay) and 10 times (screening device) for each sex, and preference indexes were assessed using Welch's t-test analysis.
We also compared the responses of the adults and larvae to larval-infected pants (B.parachinensis) and intact plants.Fifteen third instar P. xylostella larvae were placed on leaves of a B. parachinensis plant.Plants with no P. xylostella larvae (healthy plants) were used as a control.After 12 h of infestation, larvae and feces were removed and plants were placed in sealed glass jars.The behavioral responses of adults and larvae were tested using the Y-tube and the screening device, respectively, as described above.Experiments were repeated 16 times (adults) and 10 times (larvae) for each sex.The adult and larval preference indexes were calculated and analyzed by Welch's t-test.

Collection of plant volatiles and parasitoid cuticle volatiles
Fifteen third instar P. xylostella larvae were placed on leaves of a B. parachinensis plant.Plants with no P. xylostella larvae (healthy plants) were used as a control.After 12 h of infestation, larvae and feces were removed and plants were placed in sealed glass bottles.Volatiles from healthy and P. xylostella larvae-damaged plants were collected using solid-phase microextraction fibers (50/30-μm divinylbenzene/carboxen/polydimethylsioxane) for 4 h.Twenty wasps (male and female mixed) were extracted with dichloromethane (100 μL) for 30 min.The supernatants were removed, placed in a glass tube and stored at − 80 ℃ until use.The amount of heptanal in the wasp sample was evaluated by comparing the peak area to that of an internal standard (tetradecane, 100 ng diluted in 10 μL dichloromethane).Three replicates were performed for each collection types.

Fig. 1
Fig.1PxylOR16 is expressed in all the larval and adult stages of Plutella xylostella and is specifically tuned to heptanal.A Tissue expression patterns of P. xylostella OR genes.The cDNA templates for PCR analyses were from larval heads (first, second, third female, third male, fourth female, and fourth male instar larvae) and adult antennae (male adults: MA, female adults: FA).W water control.Among the 54 P. xylostella ORs, only PxylOR16 and PxylOR27 were detected in all larval stages and in adults.Actin was used for cDNA quality control.B Inward current responses of PxylOR16/PxylOrco Xenopus oocytes to plant volatile compounds (10 -4 M).C No ligand was identified for PxylOR27/PxylOrco.D Inward current responses of PxylOR16/PxylOrco Xenopus oocytes stimulated with a range of heptanal concentrations.E Response profile of PxylOR16/PxylOrco Xenopus oocytes to a panel of 71 odorants (n = 6).F Dose-response curve of PxylOR16/PxylOrco Xenopus oocyte responses to heptanal.Heptanal EC 50 = 1.757 × 10 -5 M. Error bars indicate SEM (n = 6)

Fig. 2
Fig. 2 PxylOR16 knockout mutants show impaired electrophysiological responses to heptanal.A Schematic diagram of the sgRNA target in Exon II of PxylOR16.The target sequence is shown in blue, the PAM sequence is marked in red, and the non-homologous insertion in the genome in yellow.Deleted bases are represented by dashes.PxylOR16 mutants show impaired electrophysiological responses to heptanal.B Electrophysiological responses measured as electroantennograms of Plutella xylostella antennae to heptanal and trans-2-hexen-1-ol in wild-type (WT) animals and in PxylOR16 knockout mutants generated by CRISPR/Cas9 (PxylOR16 −/−).Left, dose-dependent electroantennographic (EAG) responses of female and male moths.Heptanal was used at doses ranging from 10 ng to 10 μg.WT female and male antennae exhibited dose-dependent EAG responses to heptanal, with responses increasing with increasing heptanal doses.EAG responses of female and male PxylOR16 −/− moths to heptanal were far lower than those of WT female and male moths at all heptanal doses.Right, as a control, we tested the responses of female and male moths to trans-2-hexen-1-ol, which is not a ligand of PxylOR16.As expected, we observed no differences in the responses to this compound between PxylOR16 −/− moths and WT moths.Error bars indicate SEM (n = 10).Different letters indicate significant differences among insects (two-way ANOVA followed by Tukey's pairwise test; P < 0.05)

Fig. 4
Fig.4 Heptanal does not participate in the direct interaction between Plutella xylostella and its host plant.Preference index of P. xylostella wild-type (WT) and PxylOR16-knockout (PxylOR16 −/− ) adults and larvae to heptanal.A Preference index of WT and PxylOR16 −/− mutant female and male moths for blank (one side blank vs an intact healthy Brassica parachinensis plant), for control (an intact healthy B. parachinensis plant vs an intact healthy B. parachinensis plant), and for an intact healthy B. parachinensis plant + paraffin oil vs an intact healthy B. parachinensis plant + heptanal in a Y-tube olfactometer (n = 16).WT and PxylOR16 −/− mutant female and male moths were significantly attracted to healthy B. parachinensis plant, compared with blank.Three doses (100 ng, 1 μg, and 10 μg) and two doses (1 μg and 10 μg) of heptanal elicited avoidance behavior in WT female and male P. xylostella adults, (Welch's t-test; NS, no significant difference, P > 0.05; *** P < 0.001).The numbers of PxylOR16 −/− adults on each side of the Y-tube olfactometer were not significantly different at all heptanal doses tested, when comparing B. parachinensis + heptanal with B. parachinensis plant + paraffin oil (Welch's t-test; NS, no significant difference, P > 0.05).B Preference indexes of 10 WT and mutant female and male third instar larvae for blank (one side blank vs an intact healthy B. parachinensis plant), for control (an intact healthy B. parachinensis plant vs an intact healthy B. parachinensis plant), and for an intact healthy B. parachinensis plant + paraffin oil vs an intact healthy B. parachinensis plant + heptanal in a screening device (n = 10).Both sexes of WT and mutant larvae were attracted to the B. parachinensis plant.WT female and male larvae (third instar larvae) preferred the healthy plant with paraffin oil to a healthy plant with heptanal at three doses (100 ng, 1 μg, and 10 μg), unlike PxylOR16 −/− larvae (Welch's t-test; NS, no significant difference, P > 0.05; *** P < 0.001).C GC-MS analysis of the volatiles produced by larval-infested plants and healthy plants; heptanal was not detected.Larval-infested B. parachinensis (top), healthy B. parachinensis (middle), and synthetic heptanal (bottom).D Preference index of WT and PxylOR16 −/− mutant adults for control (intact healthy B. parachinensis plant vs intact healthy B. parachinensis plant) and for an intact healthy B. parachinensis plant vs a larval-infected B. parachinensis plant in a Y-tube olfactometer (n = 16).There was no difference in selection between the WT and PxylOR16 −/− mutant; both preferred the larval-infected plant (Welch's t-test; NS, no significant difference, P > 0.05; *** P < 0.001).E Preference index of WT and PxylOR16 −/− mutant larvae for control (intact healthy B. parachinensis plant vs intact healthy B. parachinensis plant) and for an intact healthy B. parachinensis plant vs a larval-infected B. parachinensis plant in a screening device (n = 10).Like adults, WT and PxylOR16 −/− mutant larvae both preferred the larval-infected plant (Welch's t-test; NS, no significant difference, P > 0.05; *** P < 0.001)

Fig. 5
Fig. 5 Plutella xylostella larvae and adult detect and avoid Cotesia vestalis odorants.A GC/MS analysis of the volatiles produced by C. vestalis.Heptanal could be detected in extracts of C. vestalis.IS internal standard.B Synthetic heptanal.C Preference index of WT and PxylOR16 −/− P. xylostella mutant adults (left) and larvae (right) for filter paper + dichloromethane vs filter paper + C. vestalis body wash in a Y-tube olfactometer and a 10-cm diameter plastic Petri dish, respectively (n = 16, adults; n = 28, larvae).Both sexes of WT larvae and adults of P. xylostella had avoidance responses to a C. vestalis body wash, but the PxylOR16 −/− mutant did not (Welch's t-test; NS, no significant difference; P > 0.05; * P < 0.05; *** P < 0.001).D Preference index of WT and PxylOR16 −/− mutant adults (left) and larvae (right) for air + 20 live C. vestalis vs air in a Y-tube olfactometer and a screening device, respectively (n = 16, adults; n = 10, larvae).Both sexes of WT larvae and adults of P. xylostella showed avoidance of C. vestalis, but the PxylOR16 −/− mutants did not (Welch's t-test; NS, no significant difference; P > 0.05; * P < 0.05; ** P < 0.01)