Electroantennogram response of the parasitoid, Microplitis croceipes to host-related odors: The discrepancy between relative abundance and level of antennal responses to volatile compound

Herbivores emit volatile organic compounds (VOCs) after feeding on plants. Parasitoids exploit these VOCs as odor cues to locate their hosts. In nature, host-related odors are emitted as blends of various compounds occurring in different proportions, and minor blend components can sometimes have profound effects on parasitoid responses. In a previous related study, we identified and quantified VOCs emitted by cotton plant-fed Heliothis virescens (Lepidoptera: Noctuidae) larvae, an herbivore host of the parasitoid Microplitis croceipes (Hymenoptera: Braconidae). In the present study, the olfactory response of female M. croceipes to synthetic versions of 15 previously identified compounds was tested in electroantennogram (EAG) bioassays. Using M. croceipes as a model species, we further asked the question: does the relative abundance of a volatile compound match the level of antennal response in parasitoids? Female M. croceipes showed varying EAG responses to test compounds, indicating different levels of bioactivity in the insect antenna. Eight compounds, including decanal, 1-octen-3-ol, 3-octanone, 2-ethylhexanol, tridecane, tetradecane, α-farnesene and bisabolene, elicited EAG responses above or equal to the 50 th percentile rank of all responses. Interestingly, decanal, which represented only 1% of the total amount of odors emitted by cotton-fed hosts, elicited the highest (0.82 mV) EAG response in parasitoids. On the other hand, ( E)-β-caryophyllene, the most abundant (29%) blend component, elicited a relatively low (0.17 mV) EAG response. The results suggest that EAG response to host-related volatiles in parasitoids is probably more influenced by the ecological relevance or functional role of the compound in the blend, rather than its relative abundance.


Introduction
Infested plants emit volatile organic compounds (VOCs) as an indirect defense against herbivore damage 1,2 . Informative volatile cues used by parasitoids for host location can be emitted by plants infested with herbivores 1,2 or emitted by herbivores that fed on plants 3,4 . Although plant volatiles may initially lead parasitoids to the host patch, herbivore host-specific odors are important shortrange cues used in the later stages of host location 5 . The specific mechanism by which plant-fed host larvae emit these volatiles is not fully understood. However, it is evident that parasitoids use these plant-associated VOCs in the host location process 5 . Such odor cues are usually released as a blend of various compounds in nature. Consequently, differentiating useful cues from ecologically irrelevant odors can be challenging for foraging parasitoids. Therefore, it is expected that antennal sensitivity of parasitoids will vary in response to different compounds. Antenna sensitivity in insects can be measured with electroantennogram (EAG) recording. EAG measures the summed activity of olfactory receptor neurons in the antenna and indicates the level of biological activity elicited by various compounds.
Microplitis croceipes (Hymenoptera: Braconidae) is an endoparasitoid of Heliothis virescens (Lepidoptera: Noctuidae), which is an important pest of cotton plant. In a previous related study 5 , female M. croceipes showed attraction to the odor blend emitted by cotton-fed H. virescens larvae in Y-tube olfactometer bioassays 5 . In order of elution during gas chromatography.
2 Compounds that were not tested in the present study.

Amendments from Version 1
General comment: We thank all the referees for the important comments they provided while reviewing this manuscript. In several instances, we made appropriate changes to the new version according to the suggestions of the referees. In a few other areas, we have avoided the inclusion of information that is beyond the scope of the study so as to keep the article focused and concise. As clearly stated, the present study builds on the results of a previous study by the authors. Introduction: We added a few sentences to better introduce the concept of plant-associated odors emitted by cotton-fed larvae. We clarified that odors emitted from plants play a key role in the location of host patch by parasitoids, and that odors emitted by plant-fed herbivores may be more useful in short-range host location. Finally, we added some more details about the results of the previous study conducted by the authors to create a better background for the present study.

Materials and methods:
We have included justification for the mass/volume concentration used to formulate the synthetic compounds and the single dose used in EAG recordings. Furthermore, we clarified the arrangement used to present compounds to each insect replicate. We also clarified that parasitoids used in EAG recordings were reared on host larvae that fed on artificial diet so that their sensitivity to plant odors will not be biased. Results: EAG response to solvent control was not shown because the data represent absolute EAG, in which the response to control has been deducted from responses to treatment compounds. The dataset is available for review. Discussion: The suggested citation by one of the referees has been added to the discussion. Also, we added a statement to note that parasitoids can learn to respond to diverse odor cues.

REVISED
The blend components were identified and quantified using gas chromatography-mass spectrometry (GC/MS). Furthermore, the compounds in the attractive blend occurred in varying proportions (Table 1). However, the relative abundance of a blend component does not necessarily indicate its relevance to resource location in insects 6 . In the present study, olfactory response of M. croceipes to synthetic versions of 15 previously identified compounds was tested in EAG bioassays. Comparing EAG results in the present study and GC/MS analyses in a previous study 5 , we indicated the discrepancy between relative abundance of a volatile blend component and the level of antennal response in parasitoids.

Methods and materials Insects
Microplitis croceipes was reared on 2 nd -3 rd instar larvae of Heliothis virescens. EAG recording EAG responses of M. croceipes to 15 synthetic compounds (Table 1), previously identified in the headspace of cotton-fed H. virescens larvae 5 , were recorded according to the method described by Ngumbi et al. 8 with modifications. Two compounds, α-bergamotene (not commercially available) and an unidentified compound reported in the previous study 5 were not tested in the present study. α-Pinene, β-pinene, myrcene, limonene, 2-ethylhexanol, tridecane, (E)-β-caryophyllene, α-humulene, α-farnesene and α-bisabolol with purity 95-99% were purchased from Sigma-Aldrich ® (St. Louis, MO, USA). 1-Octen-3-ol, 3-octanone, decanal, tetradecane and bisabolene with purity 96-99% were purchased from Alfa Aesar ® (Ward Hill, MA, USA). Test compounds were formulated in hexane at 0.1 μg/μl and delivered onto Whatman ® No.1 filter paper strips at an optimum dose of 1 μg. Mass/ volume concentration was used to correct for differences in purity of synthetic compounds. The dose was selected as ecologically relevant based on GC/MS analyses results of total amount of volatiles emitted by cotton-fed H. virescens larvae 5 . Impregnated filter papers were placed inside glass Pasteur pipettes and stimulus was introduced as 0.2 s odor puffs. A glass capillary reference electrode filled with 0.1 M KCl was attached to the back of the wasp head, and a similar recording electrode was connected to the excised tip of the wasp antenna. The analog signal was detected through a probe and processed with a data acquisition controller (IDAC-4, Syntech, The Netherlands). Data was assessed using EAG 2000 software (Syntech, The Netherlands). EAG responses to the 15 compounds and control (hexane) were sequentially recorded for each of 15 insect replicates. Each compound was presented at positions 1 through 15 across replicates to minimize positional bias. For instance, 1-octen-3-ol and 3-octanone were introduced to the first insect as the 3 rd and 4 th compounds, respectively, but introduced to the second insect as the 4 th and 5 th compounds, respectively.

Data analyses
Differences in absolute EAG values (EAG response to compound minus response to solvent control) of synthetic compounds were analyzed using the Kruskal-Wallis test, followed by Sidak's multiple comparison test. The relationship between EAG response and relative abundance was analyzed with Proc Corr (correlation) procedure in SAS. All analyses were performed in SAS v9.2 (SAS Institute Inc., Cary, NC, USA) with P=0.05 level of significance.

Discussion
EAG responses of Micropiltis croceipes in the present study indicated variation in biological activity elicited by test compounds at the peripheral level, and revealed a discrepancy between relative abundance and level of antennal responses in parasitoids. High EAG response elicited by decanal in M. croceipes agrees with previous reports on olfactory responses of the parasitoids, Microplitis mediator 9 and Bracon hebetor 10 . Furthermore, decanal is a key attractant for host-seeking M. croceipes 5 . Although compounds are emitted in different quantities in natural blends, minor components can have a profound effect on resource location in parasitoids 6,11 . Interestingly, decanal constituted only 1% of the total blend emitted by cotton-fed H. virescens 5 , but elicited the highest EAG response in M. croceipes, supporting the "little peaks-big effects" concept 6 . On the other hand, (E)-βcaryophyllene, the most abundant blend component, elicited a relatively low EAG response in parasitoids.
Therefore, it is more likely that the ecological relevance of a compound, rather than its relative abundance determines the level of olfactory response in foraging insects. For instance, small amounts of isothiocyanates in the volatile blend of brassica plants serve as host location cues for parasitoids of brassica herbivores 12,13 . More importantly, blend components act in concert to provide parasitoids with complete information 14 . Consequently, certain compounds function as background odors to enhance detectability (olfactory contrast) of other attractive components in a blend 12,15 . It is possible that (E)-β-caryophyllene serves as a background odor in the blend emitted by cotton-fed H. virescens. Finally, it should be noted that while EAG measures the level of bioactivity, behavioral bioassays are usually needed to establish the functional role of various compounds 5,16 . In addition, several species of parasitoids can be conditioned to respond to diverse odor cues, regardless of the relevance of such odor cues to their ecology.  Figure 1 legend. Raw data behind the representation shown in Figure 1 and analyses referred to in the Results section are included. DOI: 10.5256/ f1000research.10104.d143446 17 .
Author contributions TM and HF conceived the study. TM designed the experiment. TM and MB carried out the research. All authors contributed to writing and revision of the manuscript.

Competing interests
No competing interests were disclosed.

Grant information
The author(s) declared that no grants were involved in supporting this work.

5.
Introduction: Please give more information about why the authors specifically stated that the Heliothis larvae were cotten-fed in the lab. Will different food sources affect the compounds released virescens from the insect bodies?
Method: Please justify why mass concentration was used in preparing the compounds. Different chemicals possess various molecular weight and vapor pressure. Therefore, the number of molecules delivered onto the antenna may be dramatically different. In addition, only female wasps were used in the experiments. Apparently females need to find host to lay eggs. Just curious to know what the male's EAG responses to these compounds or if there are any related studies.
Results: Since 50% of the EAG responses from blend volatiles were used as a standard to make comparison, it would be better to add the EAG response to the blend volatiles in the bar figure. In addition, please specify the EAG response to the control solvent (hexane).
Discussion: The authors initiated a good start to discuss some compounds in the blend may function as background odors to enhance olfactory contract. There are many excellent reviews about the possible mechanisms behind this pheromone, such as Riffell and Hildebrand (2016).The authors may discuss a little bit about the mechanisms.
No competing interests were disclosed. Competing Interests: The wasps in your study had the experience of living and hatching from larvae having fed on cotton. Thus 1 1 The wasps in your study had the experience of living and hatching from larvae having fed on cotton. Thus they might have experience with the compounds you present in Table 1. It has been shown that M.

I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.
can learn almost any compund (e.g. Olson et al. ). croceipes In your discussion you point out that it might be the ecological relevance of a compound that determines the antennal response -however, Park et al. showed in electroantennogram studies that the antenna of is also responding to anthropogenic compounds with high sensitivity. Thus, the antennal M. croceipes response might not reflect the ecological relevance. This might be more reflected in the behavioural response, as you indicate in your discussion. And this might be fine-tuned by leaning in case of a parasitoid with a polyphagous host.

Minor points:
Methods: Why is 1 µg an optimal dose? The manuscript reported the the relationship between the relative abundance of volatile chemical emitted from larvae and the antennal response by the larval parasitoid of the herbivore, Heliothis virescens . By electroantennogram bioassays, the authors found that the level of parasitoid's Microplitis croceipes EAG response is not related to relative abundance of the volatile components in the blend. Since a chemical that elicits an EAG response in an insect does not mean to elicit a behavioral response, and a chemical eliciting a bigger EAG response does not mean to elicit a stronger behavioral response, these experiments seem little and the novelty and significance of this study seem limited. Introduction: In a previous study, the authors have investigated the attractiveness of these individual volatile compounds to the parasitoid, thus it would be better to introduce these results briefly in the section of Introduction.
Methods: It has been well documented that an insect has different responsive ranges to different chemicals. Thus, only using one concentration of chemicals is not enough.
Discussion: Based on the previous results reported by the authors , 8 chemicals in the blend, including decanal and ( )-β-Caryophyllene, had a role in attraction of the parasitoid. The authors should give a E discussion based on above results.

Specific comments
The title is appropriate for the content of the article; however, it can be made more concise. E.g. the Title: first part of the existing title would adequate, i.e. 'Electroantennogram response of the parasitoid, to host-related odors'.

Microplitis croceipes
The introduction clearly states the objective of the study. However, adequate background Introduction: is missing in the area of herbivore emitted plant associated VOCs. Are these plant derived volatile compounds emitted by herbivore itself (after feeding) or are they adsorbed into the hervivore body during feeding process (e.g. from frass) The authors followed standard insect rearing (Lewis & Burton, 1970) and Methods and materials: electroantennogram recording (Ngumbi , 2010) and data analysis procedures. However, the number et al. of insect replicates used in the study is not clear. Was a single insect antennal preparation used to record EAG responses to the 15 compounds and control (hexane)? Was the abundance of volatile components varied based on the corresponding quantities in the natural headspace samples? The latter has not been specified in the methodology except mentioning 'Test compounds were formulated in hexane at 0.1 μg/μl '. and delivered onto Whatman®No.1 filter paper strips at an optimum dose of 1 µg The results show that female showed varying EAG responses to test compounds.

Results:
M. croceipes Notably, decanal which constituted only 1% of total blend emitted by cotton-fed elicited the H. virescens highest EAG response (0.82 mV); while ( )-β-caryophyllene, the most abundant component (29.2% of E total blend), elicited a relatively low EAG response (0.17 mV) in the parasitoid antenna. This is possible as earlier reports also indicated compounds with highest EAG response may not necessarily be those emitted in largest quantity (Tamiru 2015 ). Given the fact that the main research question of this et al. study is examine the relationship between relative abundance and level of antennal responses, it would have been more informative to test different concentrations/amounts of the test compounds and their corresponding EAG response to reliably measure statistical significance of correlations between relative abundance of the compounds and level of EAG response.
The authors discuss about the ecological relevance of a compounds. However, Discussion: electrophysiological responses do not necessarily mean that a behavioral response will occur; rather it elucidates potential behaviorally relevant compounds. To fully explore the significance of the results, bioassays need to be carried out with identified compounds, both individually and as a blend, to determine the kind of behavioral response the volatile compounds trigger in the parasitoid and their ecological/functional role. I suggested authors to refer and perhaps include in their discussion the work by Tamiru et al. (2015) which demonstrated combined use of electrophysiological and behavioral studies for better understanding of odor mediated behavior in insects.
I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.