1‐Undecene from Pseudomonas aeruginosa is an olfactory signal for flight‐or‐fight response in Caenorhabditis elegans

Abstract Animals possess conserved mechanisms to detect pathogens and to improve survival in their presence by altering their own behavior and physiology. Here, we utilize Caenorhabditis elegans as a model host to ask whether bacterial volatiles constitute microbe‐associated molecular patterns. Using gas chromatography–mass spectrometry, we identify six prominent volatiles released by the bacterium Pseudomonas aeruginosa. We show that a specific volatile, 1‐undecene, activates nematode odor sensory neurons inducing both flight and fight responses in worms. Using behavioral assays, we show that worms are repelled by 1‐undecene and that this aversion response is driven by the detection of this volatile through AWB odor sensory neurons. Furthermore, we find that 1‐undecene odor can induce immune effectors specific to P. aeruginosa via AWB neurons and that brief pre‐exposure of worms to the odor enhances their survival upon subsequent bacterial infection. These results show that 1‐undecene derived from P. aeruginosa serves as a pathogen‐associated molecular pattern for the induction of protective responses in C. elegans.

11th Nov 2020 1st Editorial Decision Dear Dr. Singh, Thank you for submitting your manuscript to The EMBO Journal. Your study has now been seen by three referees and their comments are provided below.
As you can see from the comments, the referees find the analysis interesting. However significant revisions are also needed in order to consider publication here. Important controls are missing, some of the data needs to be better substantiated and further insight into how the induction of the host immune response pathway is linked to survival is needed. Should you be able to extend the findings along the lines suggested by the referees then I would be willing to consider a revised version.
I am happy to discuss the raised points further and maybe it would be most helpful to do so via phone or video.
When preparing your letter of response to the referees' comments, please bear in mind that this will form part of the Review Process File, and will therefore be available online to the community. For more details on our Transparent Editorial Process, please visit our website: https://www.embopress.org/page/journal/14602075/authorguide#transparentprocess We generally allow three months as standard revision time. As a matter of policy, competing manuscripts published during this period will not negatively impact on our assessment of the conceptual advance presented by your study. However, we request that you contact the editor as soon as possible upon publication of any related work, to discuss how to proceed. Should you foresee a problem in meeting this three-month deadline, please let us know in advance and we may be able to grant an extension.
Thank you for the opportunity to consider your work for publication. I look forward to discussing your revisions further with you Yours sincerely, Karin Dumstrei, PhD Senior Editor The EMBO Journal Instructions for preparing your revised manuscript: Please make sure you upload a letter of response to the referees' comments together with the revised manuscript.
IMPORTANT: When you send the revision we will require -a point-by-point response to the referees' comments, with a detailed description of the changes made (as a word file).
-a word file of the manuscript text.
-individual production quality figure files (one file per figure) -a complete author checklist, which you can download from our author guidelines (https://www.embopress.org/page/journal/14602075/authorguide). -Expanded View files (replacing Supplementary Information) Please see out instructions to authors https://www.embopress.org/page/journal/14602075/authorguide#expandedview Please remember: Digital image enhancement is acceptable practice, as long as it accurately represents the original data and conforms to community standards. If a figure has been subjected to significant electronic manipulation, this must be noted in the figure legend or in the 'Materials and Methods' section. The editors reserve the right to request original versions of figures and the original images that were used to assemble the figure.
Further information is available in our Guide For Authors: https://www.embopress.org/page/journal/14602075/authorguide The revision must be submitted online within 90 days; please click on the link below to submit the revision online before 9th Feb 2021. In this manuscript, the authors report that a Pseudomonas specific volatile odor elicits behavioral avoidance and induces immune-specific gene expression, increasing survival in response to pathogenic bacteria. There is evidence that host immune responses may be driven by chemosensory cues in multiple systems, although the molecular mechanisms of these responses are largely unknown. Recently, it was demonstrated that another C. elegans olfactory neuron -AWC -plays a critical role in non-autonomous regulation of p38 MAPK dependent immune responses via unidentified odors (Foster et al., 2020). Thus, the identification of a parallel olfactory input, including the identity of both the odors and molecular pathways involved, would be a really novel finding of broad interest.
The authors show that 1-undecene elicits behavioral avoidance and induces immune response genes and that undecene production by Pseudomonas is necessary for these responses. This is a really interesting finding, however, at this point there are many cases in which missing controls and/or an insufficiently fleshed out pathway severely limits the interpretation of these data. Additionally, it is unclear how the innate immune responses shown here are coordinated with the observed acute behavioral responses to alter survival.
Specific comments: 1. A major concern is the lack of both genetic and chemical controls for most of the experiments. For all of the C. elegans mutants, only a single allele was analyzed and no rescue experiments were performed. This is important to not only control for genetic background, but also to confirm the specificity of the proposed AWB neuron function. For example, odr-3 is expressed in AWB and AWC, as noted by the authors, but also in ASH, where it is necessary for odor responses (see Yoshida et al., 2012). How specific are the observed responses for 1-undecene? Can the authors use an analogous volatile chemical that does not activate AWB in the imaging and gene induction experiments?
2. The effects on foraging behavior are interesting but are a bit confusing. Do these locomotory changes lead to increased lawn-leaving behavior? The increased survival on a partial lawn of PA14 suggests they should.
3. How does the 1-undecene dependent induction of innate immune response genes relate to the survival of the worms on PA14? The authors propose that 1-undecene acts via the zip-2 pathway to induce irg-1/2/3 expression. zip-2 mutants have decreased survival on PA14 lawns, but the authors only observed a survival effect on partial lawns. Does this mean the immune response induction does not alter survival and is secondary to the behavioral response? Is survival reduced on undA mutant bacteria and does undA play a role in virulence? A major question in the field is how chemosensory and innate immune reponses are coordinated to alter survival and in my opinion, this could be further fleshed out. 4. Gene expression experiments: for the RT-PCR experiments, it is unclear how many replicates were performed or how the experiments were designed. For example, it appears that Figure 1A and 1C contain the same data for irg-1/2/3, if this is the case it should be explicitly noted. If Figure 1A was essentially exploratory in nature, then in my opinion, the control data should have been replicated in order to test a different hypothesis in 1C. An alternative might be to use reporters, as in 4B, but including the appropriate controls to demonstrate an AWB specific effect. Similar controls should be performed to compare the wild-type vs. undA mutants.
Minor suggestions: 1. The chemotaxis effect in the bacterial undA mutant are striking, and it seems odd that these data are not included in the main figures. 2. There are several grammatical errors in the manuscript which could use a little more proofreading. 3. I could not find a methods entry for the acute avoidance experiments and what, if any, controls were done for this experiment? 4. Figure 3C-F all seem to show essentially the same data, perhaps this could be made more concise? 5. Line 292 makes reference to "...aversion response in at least 80% of the population" -it's unclear to what this is referencing.
Referee #2: In this study "1-Undecene from P. aeruginosa is an olfactory signal for flight or fight response in C. elegans", the authors identified 1-undecene as a volatile product of a type of infectious Pseudomonas aeruginosa. They showed that 1-undecene was repulsive to C. elegans worms and Pseudomonas aeruginosa unable to produce 1-undecene could not repel C. elegans. 1-undecene stimulates GCaMP3 signals in AWB neurons, and repulsion of 1-undecene to C. elegans is lost in a lim-4 mutant. Exposure to 1-undecene induces expression of several zip-2-dependent immune response genes. Overall, the authors aim to address an interesting question. The experiments are well designed, and the results are likely to contribute to a better understanding of host-pathogen interactions. Addressing or clarifying the following several questions will help to improve the manuscript.
1. The authors used an odr-3 mutant to examine the function of odor sensation in their analysis. In addition to odor sensory neurons, odr-3 is also expressed in other chemosensory neurons including nociceptive neurons ASH, phasmid neurons PHB. The results based on testing odr-3 are not fully conclusive about the involvement of odor sensation. It is helpful that authors used a lim-4 mutant to strengthen the conclusion. However, lim-4 is found in many neurons including motor neurons. The movement defects of the lim-4 mutant may interfere with chemotaxis assays. An AWB-specific ablation line or a similar reagent is needed to support the function of AWB in this study. 2. Because one single mutant was used for odr-3 and for lim-4, the conclusion based on the current results needs to be confirmed with rescue. Or, the authors should at least test a second mutation for each of these genes. 3. Figure 2b shows that when 1-undecene is tested alone, it repels worms. However, it is not clear whether 1-uncedene repels worms when it is a part of the volatile products of a bacterial lawn. The authors should test whether 1-undecene added to a lawn of E. coli repels worms and quantify the lawn leaving behavior similarly as in figure 1a and 1b. In figure 2d the authors showed trajectories of worms. However, it is difficult to see on these panels where 1-undecene is and where the lawn borders are. In addition, the results need to be quantified. 4. "We observed increased speed and enhanced roaming by worms in an arena exposed to 1undecene odor compared to the arena without the repellent ( Fig. 2D and S2B). Worms also executed omega bend in response to 1-undecene in 16 out of 17 worms ( Figure 2C, Supplementary Movie S3) consistent with response to an aversive signal." The speed and the omega bend need to be quantified and compared with controls. 5. T-test is perhaps not the best statistical test for the data in fig 1e, 3b and 3d. 6. Is 1-undecene a microbe-associated molecular pattern or a pathogen-associated molecular pattern? The authors propose the later, because exposure to 1-undecene induces expression of zip-2-dependent immune genes. If this is the case, mutants of PA14 with weaker virulence would not produce 1-undecene. Did author test this possibility? Clarification of this question will better characterize the property and function of 1-undecene. 7. "Moreover, we also could not detect 1-undecene in E. coli OP50, Salmonella typhimurium, and Enterococcus faecalis headspace (data not shown)". It will be very helpful to include these data. 8. It is intriguing that exposure to 1-undecene specifically upregulates zip-2-dependent immune genes irg-1,2,3. Does the induction depend on zip-2? 9. The logic for line 47-65 should be better clarified. The authors summarized several recent studies that addressed fight (immune response) and flight (avoidance and learning) responses to pathogens, odorant sensation of pathogens, virulence-independent sensation, and anatomical changes caused by pathogen exposure. It is useful for the authors to better organize these information and clarify how some or all of these studies relate to their question "However, the nature of P. aeruginosa volatile molecules that facilitate pattern recognition and threat perception in C. elegans remain unclear." 10. "However, worms have a large sensory repertoire of 302 neurons..." This sentence is not accurate.

Referee #3:
Through an analytical chemistry approach, Prakash et al. identify 1-undecene as a volatile produced by P. aeruginosa that can alter behavior and gene expression in the nematode C. elegans. They present data suggesting that it acts via the neuron AWB. The responses of C. elegans to pathogens remains an area of active research and these results will potentially interest those working in the field. Most of the data appears robust and generally the study is well presented. There are some recent studies that should have been cited, including Ringstad's on O2/CO2 sensing, and most pertinently, "Identification of Odor Blend Used by Caenorhabditis elegans for Pathogen Recognition", Worthy et al. (2018) in the specialist journal, Chemical Senses.
The most significant claim that Prakash et al. make, "1-undecene serves as a molecular pattern and induces upregulation of a subset of immune response genes specific to P. aeruginosa in worms", is currently not sufficiently supported by the data. To demonstrate that 1-undecene has a specific effect on host gene expression, indicative of a pathogen-specific immune response, they assayed the expression of genes that are induced by P. aeruginosa, including irg-1 and irg-2, and genes that are not induced by P. aeruginosa. Work from the Troemel lab and others, has shown that irg-1 and irg-2 are far from being pathogen-specific markers. They have been suggested to be part of an "Intracellular Pathogen Response" genes (Reddy et al 2017), regulated by pals-22 and pals-25 (Reddy et al. 2019), and induced by a range of different pathogens (e.g Orsay virus for irg-2), as a consequence of cellular dysfunction caused by infection. This is compatible with the fact irg 1 and irg 2 expression is strongly induced by the translational elongation inhibitor cycloheximide (Dunbar 2012). 1-undecene is far for a biologically inert molecule; it is known to have antimicrobial activity for example (10.3389/fmicb.2015.01082 and references therein). Its effects on host gene expression could reflect disruption of cellular homeostasis. If the authors want to support their suggestion of a pathogen-specific response, they need to demonstrate that the 1-undecene-associated increases in gene expression are independent of pals regulation. They should also examine the expression of different markers of cellular stress. Further, they should assay the expression of candidate genes that they recently identified as being regulated by P. aeruginosa, E. faecalis and C. neoformans, such as dct-17 and lys-3.
Other minor concerns that should be addressed: Figure 1F: N2 appear to have 2 distinct populations, one responsive and one un-responsive. There is a similar separation in Figure S2D. What is the explanation?
Welch's t-test is insensitive to equality of the variances, but does assume normal distribution. It should not be used for non-normal data (like Figure 1F).
If the control data for Figure 4C is identical to the data in Figure 4A, this needs to be indicated.
qRT-PCR is presented, for example, as "Relative mRNA expression N2 (1-Undecene / Naive)", but there are negative values, so this cannot be right. What is actually being represented?
The authors make a distinction between "younger" and "older" lawns. If 1-undecene is not produced by P. aeruginosa in liquid culture and/or at 37C, the differences observed would simply be a question of the time that the bacterial cultures have been in a situation compatible with 1undecene synthesis. This could readily checked by assaying undA expression under the different culture conditions.
In this manuscript, the authors report that a Pseudomonas specific volatile odor elicits behavioral avoidance and induces immune-specific gene expression, increasing survival in response to pathogenic bacteria. There is evidence that host immune responses may be driven by chemosensory cues in multiple systems, although the molecular mechanisms of these responses are largely unknown. Recently, it was demonstrated that another C. elegans olfactory neuron -AWC -plays a critical role in non-autonomous regulation of p38 MAPK dependent immune responses via unidentified odors (Foster et al., 2020). Thus, the identification of a parallel olfactory input, including the identity of both the odors and molecular pathways involved, would be a really novel finding of broad interest.
The authors show that 1-undecene elicits behavioral avoidance and induces immune response genes and that undecene production by Pseudomonas is necessary for these responses. This is a really interesting finding, however, at this point there are many cases in which missing controls and/or an insufficiently fleshed out pathway severely limits the interpretation of these data.
Additionally, it is unclear how the innate immune responses shown here are coordinated with the observed acute behavioral responses to alter survival.
Specific comments: 1. A major concern is the lack of both genetic and chemical controls for most of the experiments.
For all of the C. elegans mutants, only a single allele was analyzed and no rescue experiments were performed. This is important to not only control for genetic background, but also to confirm the specificity of the proposed AWB neuron function. For example, odr-3 is expressed in AWB and AWC, as noted by the authors, but also in ASH, where it is necessary for odor responses (see Yoshida et al., 2012). How specific are the observed responses for 1-undecene?
Can the authors use an analogous volatile chemical that does not activate AWB in the imaging and gene induction experiments?

RESPONSE:
We agree with the reviewer that ODR-3 is also expressed non-olfactory neurons such as nociceptive neuron ASH which can respond to odors (Yoshida et al., 2012). In the revised manuscript, we have used additional allele for odr-3(n2046), in addition to odr-3(n2150) as well as lim-4(yz12) in addition to lim-4(ky403). These experiments are described in response to the next comment. We have also used ablation for ASH neurons to show that this neuron is 13th Feb 2021 1st Authors' Response to Reviewers not involved in response to 1-undecene ( Figure 3B). Additionally, we also analysed GCaMP response of ASH neurons to 1-undecene and found no response (Rebuttal Figure 1).
To control for volatile chemical stimuli, we have studied the transcriptional response of C. elegans upon 2 hours exposure to diacetyl (2,3 butanedione), an odor sensed by AWA olfactory neurons. We found no significant upregulation of irg-1, irg-2 or irg-3 (shown in rebuttal Figure 2 A and 2B) by qRT-PCR or using irg-1::GFP reporter..

Rebuttal Figure 1:
Heat map of calcium response in ASH::GCaMP3. Each row represents an individual worm recorded for 180 s under 1:100 dilution of 1-undecene, in 11 s-130 s window.
2. The effects on foraging behavior are interesting but are a bit confusing. Do these locomotory changes lead to increased lawn-leaving behavior? The increased survival on a partial lawn of PA14 suggests they should.  RESPONSE: We believe that locomotory behavior is likely a search for area with lower concentration of 1-undecene. And yes, we agree with the reviewer that avoidance is a survival strategy for worms as our experiments with partial and full lawn and with odr-3 mutants ( Figure 1 in the revised manuscript) would indicate.
3. How does the 1-undecene dependent induction of innate immune response genes relate to the survival of the worms on PA14? The authors propose that 1-undecene acts via the zip-2 pathway to induce irg-1/2/3 expression. zip-2 mutants have decreased survival on PA14 lawns, but the authors only observed a survival effect on partial lawns. Does this mean the immune response induction does not alter survival and is secondary to the behavioral response? Is survival reduced on undA mutant bacteria and does undA play a role in virulence? A major question in the field is how chemosensory and innate immune reponses are coordinated to alter survival and in my opinion, this could be further fleshed out.

RESPONSE: Thanks for your suggestions.
We have pre-exposed worms to 1-undecene before survival assays on P. aeruginosa. As shown in Figure 4 (new panel added in Figure 4F and 4G of the revised manuscript), pre-exposure to 1undecene enhances the survival of worms from P. aeruginosa compared to naive worms. The experiment is shown as rebuttal Figure 3 here.
If 1-undecene is indeed a molecular pattern necessary for induction of protective immune response in worms, we expected to see reduced survival of worms on undA compared to on wild type P. aeruginosa PA14. As shown in rebuttal Figure 4 C. elegans survival is indeed reduced on undA mutant. Since we observed induction of irg-1, irg-2 and irg-3 within two hours of exposure, we would like to classify it as an immediate-early response.
Rebuttal Figure 3: Preexposure of worms to 1-undecene enhances the survival of worms upon subsequent infection with P. aeruginosa. (A) Kaplan Meier survival curve on P. aeruginosa for N2 (naive) worms and N2 worms pre-exposed to 1-undecene odor. Survival assay was performed at 20°C. (B) Time required for 50% of worms to die (TD50) on P. aeruginosa. Each data point indicates replicates with ~100 worms. n = 3 assays. ** P ≤ 0.01 as determined by two-tailed unpaired ttest. Error bars indicate SEM. Survival assay was performed at 20°C. (B) Time required for 50% of worms to die (TD50) on P. aeruginosa wild type (PA14) and undA mutant. Each data point indicates replicates with ~100 worms. n = 3 assays. * P ≤ 0.05 as determined by two-tailed unpaired t-test. Error bars indicate SEM.

Rebuttal
4. Gene expression experiments: for the RT-PCR experiments, it is unclear how many replicates were performed or how the experiments were designed. For example, it appears that Figure 1A and 1C contain the same data for irg-1/2/3, if this is the case it should be explicitly noted. If Figure 1A was essentially exploratory in nature, then in my opinion, the control data should have been replicated in order to test a different hypothesis in 1C. An alternative might be to use reporters, as in 4B, but including the appropriate controls to demonstrate an AWB specific effect. Similar controls should be performed to compare the wild-type vs. undA mutants.

RESPONSE:
We think the reviewer is referring to qRT-PCR analyses in Figure 4. We have taken care to describe the number of biological replicates for each experiment in the methods section of the revised manuscript. All the control data (N2 worms exposed to 1-undecene odor) has been replicated several times for the revision and included in revised Figure 4. We have ensured that control data (WT worms exposed to 1-undecene) is from different experiments across panels in Figure 4. We have also analysed the transcriptional response of 1-undecene exposure in two alleles of odr-3 and 2 alleles of lim-4 in the revised Figure 4.
Minor suggestions: 1. The chemotaxis effect in the bacterial undA mutant are striking, and it seems odd that these data are not included in the main figures.

RESPONSE:
We have included the choice assay for undA mutant (over wild type PA14) in main Figure 2 in the revised manuscript.
2. There are several grammatical errors in the manuscript which could use a little more proofreading.

RESPONSE:
We have proofread the entire manuscript.
3. I could not find a methods entry for the acute avoidance experiments and what, if any, controls were done for this experiment? RESPONSE: For chemotaxis assays, standard protocol was used with some modifications (Bargmann et al., 1993). The details are included in the Methods section and Supplementary Figure S1D of the revised manuscript. As a control, we performed chemotaxis assays against a well-studied repellent 2-nonanone and attractant diacetyl. The data for controls is shown here (rebuttal Figure 5).
Rebuttal Figure 5: Chemotactic response of N2 worms towards diacetyl and 2-nonanone. Each data point indicates an individual assay plate with ~40 worms. n ≥ 3 assays. Error bars indicate SEM.
4. Figure 3C-F all seem to show essentially the same data, perhaps this could be made more concise?
RESPONSE: Thanks for your comments. Since our study is the first report on 1-undecene as a repellent and the first report of stimulation for AWB neurons, all the panels provide useful information for the C. elegans community. Therefore, we would like to retain all the panels.
5. Line 292 makes reference to "...aversion response in at least 80% of the population" -it's unclear to what this is referencing. RESPONSE: We Have rephrased the statement for better understanding.
"A large fraction of worms in a population showed aversion response suggesting that 1undecene is a physiologically relevant stimulus for worms."  Referee #2: In this study "1-Undecene from P. aeruginosa is an olfactory signal for flight or fight response in C. elegans", the authors identified 1-undecene as a volatile product of a type of infectious Pseudomonas aeruginosa. They showed that 1-undecene was repulsive to C. elegans worms and Pseudomonas aeruginosa unable to produce 1-undecene could not repel C. elegans. 1-undecene stimulates GCaMP3 signals in AWB neurons, and repulsion of 1-undecene to C. elegans is lost in a lim-4 mutant. Exposure to 1-undecene induces expression of several zip-2-dependent immune response genes. Overall, the authors aim to address an interesting question. The experiments are well designed, and the results are likely to contribute to a better understanding of hostpathogen interactions. Addressing or clarifying the following several questions will help to improve the manuscript.
1. The authors used an odr-3 mutant to examine the function of odor sensation in their analysis.
In addition to odor sensory neurons, odr-3 is also expressed in other chemosensory neurons including nociceptive neurons ASH, phasmid neurons PHB. The results based on testing odr-3 are not fully conclusive about the involvement of odor sensation. It is helpful that authors used a lim-4 mutant to strengthen the conclusion. However, lim-4 is found in many neurons including motor neurons. The movement defects of the lim-4 mutant may interfere with chemotaxis assays. An AWB-specific ablation line or a similar reagent is needed to support the function of AWB in this study.

RESPONSE:
We agree with the reviewer that ODR-3 is expressed in non-olfactory neurons such as nociceptive neuron ASH which can respond to odors. We have used ablation for ASH neurons to show that this neuron is not involved in response to 1-undecene ( Figure 3B in the revised manuscript). We also analyzed GCaMP response of ASH neurons to 1-undecene and found no response (Rebuttal Figure 1). In the revised manuscript, we have used an additional allele odr-3(n2046), in addition to odr-3(n2150). Our efforts to create AWB ablation strain have failed so far and we have not found viable ablation lines. Therefore, we have utilized an additional allele of lim-4(yz12) in addition to lim-4(ky403) and were able to phenocopy all the phenotypes.
2. Because one single mutant was used for odr-3 and for lim-4, the conclusion based on the current results needs to be confirmed with rescue. Or, the authors should at least test a second mutation for each of these genes.
RESPONSE: Thank you. We have used additional allele odr-3(n2046), in addition to odr-3(n2150) as well as lim-4(yz12) in addition to lim-4(ky403). We have included data for both the alleles of odr-3 in Figure 1 in the revised manuscript. This includes lawn leaving data ( Figure  1A), survival assays (Figure 1 C-E). Transcriptional response for both alleles of odr-3 is included in Figure 4D in the revised manuscript.
We have included two alleles of lim-4 in 1-undecene chemotaxis assays ( Figure 3B) and on the transcriptional response to 1-undecene exposure ( Figure 4E). Both alleles fail to show induction of irg-1, irg-2 and irg-3 in response to 1-undecene.
3. Figure 2b shows that when 1-undecene is t d alone, it repels worms. However, it is not clear whether 1-uncedene repels worms when it is a part of the volatile products of a bacterial lawn. The authors should test whether 1-undecene added to a lawn of E. coli repels worms and quantify the lawn leaving behavior similarly as in figure 1a and 1b. In figure 2d the authors showed trajectories of worms. However, it is difficult to see on these panels where 1-undecene is and where the lawn borders are. In addition, the results need to be quantified.

RESPONSE:
Thanks for this suggestion. When 1-undecene was adsorbed in OP50 lawn, we could observe multiple reversals and omega turns on the lawn, as expected but we did not observe complete lawn leaving response.
We designed a different experiment to address if bacteria prefer OP5O lawn over OP50 lawn with 1-undecene. As described in the schematic in Rebuttal Figure 6, two OP50 lawns, A and B were separated by 5 mm. In test plates,1-undecene was absorbed on lawn B or left as is in control plates. Worms were placed on lawn A and we examined if worms cross over to lawn B in 2 hours. In control plates, we found that worms moved to lawn B in all the control plates but only a small fraction of worms moved to lawn B in test plates. Based on this experiment, we inferred that C. elegans prefers OP50 food without 1-undecene.
Rebuttal Figure 6: (A) Schematic of the experimental setup used to record the repulsion behavior of worms from 1undecene. Lawn A and B are E. coli OP50. Lawn B in the test condition is spiked with 0.5 µl of 1undecene (test lawn). Worms were introduced in lawn A at the start of the experiment and the distribution of worms on lawn B was recorded after an interval of 120 m in both control and test condition. (B) Percent distribution of worms in the test lawn (lawn B) in both control and test condition after an interval of 120 minutes. Each data point represents one assay plate with ~20 worms. n ≥ 3 assays. *** P ≤ 0.001 as determined by two-tailed unpaired t-test. Error bars indicate SEM.
4. "We observed increased speed and enhanced roaming by worms in an arena exposed to 1undecene odor compared to the arena without the repellent (Fig. 2D and S2B). Worms also executed omega bend in response to 1-undecene in 16 out of 17 worms ( Figure 2C, Supplementary Movie S3) consistent with response to an aversive signal." The speed and the omega bend need to be quantified and compared with controls.

RESPONSE:
We have quantified omega turns in C. elegans exposed to 1-undecene or no odor for 5 minutes. All the worms exposed to 1-undecene odor executed omega turns (rebuttal figure  7A).
We have also quantified the speed of the worms under 1-undecene odor exposure and control condition (rebuttal figure 7B). The coordinate information of worms from 2604 tracking dataset was imported into MATLAB 2019B. The imported data was then used to calculate the distance by using Euclidean distance formula. Distance obtained was converted into speed by dividing with time interval between two consecutive frames. The probability density histogram was generated (bin size of 0.05) for worms' track in the control and 1-undecene exposed conditions. Further, the significance was determined using Kolmogorov-Smirnov test. This data suggests that the average speed of worms exposed to 1-undecene was significantly higher compared to control.
Rebuttal Figure 7: (A) Percent of worms performing omega turn as observed under a window of 5 minutes. Each data point represents an average of 5 worms tested. n = 3. *** P ≤ 0.001 as determined by twotailed unpaired t-test. Error bars indicate SEM. (B) Histogram of probability density distribution of the velocity threshold of worms exposed to 1undecene odor or control condition. Significance was calculated using Kolmogorov-Smirnov test (P =0.0144). RESPONSE: Figure 1E has been analyzed with two-tailed, unpaired t-test (Ha et al., 2010;Styer et al., 2008). 6. Is 1-undecene a microbe-associated molecular pattern or a pathogen-associated molecular pattern? The authors propose the later, because exposure to 1-undecene induces expression of zip-2-dependent immune genes. If this is the case, mutants of PA14 with weaker virulence would not produce 1-undecene. Did author test this possibility? Clarification of this question will better characterize the property and function of 1-undecene.

RESPONSE:
We hypothesize that P. aeruginosa strains with apparent increased virulence may not produce (or produce less of) 1-undecene resulting in dampened host immune response. We see reduced survival of C. elegans on undA mutant (rebuttal Figure 4) suggesting that 1undecene is necessary for induction of immune response, consistent with our hypothesis. Additionally, we examined undA transcript level in a quorum-sensing defective and hypo virulent mutant rhlR. As shown in rebuttal Figure 8, we found no significant alteration in expression suggesting that UndA is not under the control of quorum. This suggests that there  7. "Moreover, we also could not detect 1-undecene in E. coli OP50, Salmonella typhimurium, and Enterococcus faecalis headspace (data not shown)". It will be very helpful to include these data.
RESPONSE: Thank you. We show data for E. coli OP50 and Enterococcus faecalis OG1RF (Rebuttal Figure 9). As shown, the 1-undecene peat at 14 minutes is not seen in either of these two bacteria. We also closely surveyed the literature for several bacteria (Staphylococcus aureus, Salmonella Typhimurium, Serratia marcescens) and found that none of these produced 1undecene (Siripatrawan, 2008;Worthy et al., 2018;Zhu et al., 2010).  8. It is intriguing that exposure to 1-undecene specifically upregulates zip-2-dependent immune genes irg-1,2,3. Does the induction depend on zip-2? RESPONSE: Yes, the transcription of irg-1 and irg-2 is dependent in ZIP-2 transcription factor. We found that 1-undecene odor exposure for 2 hours failed to induce irg-1 and irg-2 in zip-2(tm4248) mutant. This is included in the revised manuscript as Figure 4B.
9. The logic for line 47-65 should be better clarified. The authors summarized several recent studies that addressed fight (immune response) and flight (avoidance and learning) responses to pathogens, odorant sensation of pathogens, virulence-independent sensation, and anatomical changes caused by pathogen exposure. It is useful for the authors to better organize these information and clarify how some or all of these studies relate to their question "However, the nature of P. aeruginosa volatile molecules that facilitate pattern recognition and threat perception in C. elegans remain unclear."

RESPONSE:
We have reorganized this paragraph in the revised manuscript, to lead to the question of whether volatile molecules serve as microbe-associated molecular patterns. 10. "However, worms have a large sensory repertoire of 302 neurons..." This sentence is not accurate. RESPONSE: We have rephrased to: "However, worms have a large sensory repertoire of ~1300 G protein-coupled receptors."

Referee #3:
Through an analytical chemistry approach, Prakash et al. identify 1-undecene as a volatile produced by P. aeruginosa that can alter behavior and gene expression in the nematode C. elegans. They present data suggesting that it acts via the neuron AWB. The responses of C. elegans to pathogens remains an area of active research and these results will potentially interest those working in the field. Most of the data appears robust and generally the study is well presented. There are some recent studies that should have been cited, including Ringstad's on O2/CO2 sensing, and most pertinently, "Identification of Odor Blend Used by Caenorhabditis elegans for Pathogen Recognition", Worthy et al. (2018) in the specialist journal, Chemical Senses.
RESPONSE: Thank you. Several additional references have been included in the revised manuscript including those mentioned by the reviewer. We have been unable to accommodate Ringstad et al, 2013 in our study as it is focused on CO2 sensing neuron and not linked to pathogen recognition or secondary metabolite perception, to the best of our knowledge.
The most significant claim that Prakash et al. make, "1-undecene serves as a molecular pattern and induces upregulation of a subset of immune response genes specific to P. aeruginosa in worms", is currently not sufficiently supported by the data.

RESPONSE:
In the revised manuscript, we have pre-exposed worms to 1-undecene volatile followed by analysis of survival on live P. aeruginosa lawn. We find that pre-exposure enhances resistance of worms proving evidence that immune response induction is linked to protection (new panel added in Figure4, 4F and 4G in the revised manuscript). The experiment is shown as rebuttal Figure 3 here. irg-1, irg-2 and irg-3 are specific to P. aeruginosa as shown previously by Troemel lab (Estes et al., 2010).
We have also analysed 8 immune effectors specific to Gram-positive bacteria and pathogenic yeast and find that they are not upregulated by exposure to 1-undecene odor (Fig. S4D). 1undecene also does activate heat shock response, oxidative stress response (Fig. S4 G-H in the revised manuscript), or Intracellular pathogen response (rebuttal Figure 10). Collectively these experiments provide evidence that 1-undecene odor induces specific protection against P. aeruginosa and does not induce other responses.
To demonstrate that 1-undecene has a specific effect on host gene expression, indicative of a pathogen-specific immune response, they assayed the expression of genes that are induced by P. aeruginosa, including irg-1 and irg-2, and genes that are not induced by P. aeruginosa. Work from the Troemel lab and others, has shown that irg-1 and irg-2 are far from being pathogenspecific markers. They have been suggested to be part of an "Intracellular Pathogen Response" genes (Reddy et al 2017), regulated by pals-22 and pals-25 (Reddy et al. 2019), and induced by a range of different pathogens (e.g Orsay virus for irg-2), as a consequence of cellular dysfunction caused by infection. This is compatible with the fact irg 1 and irg 2 expression is strongly induced by the translational elongation inhibitor cycloheximide (Dunbar 2012). 1undecene is far for a biologically inert molecule; it is known to have antimicrobial activity for example (10.3389/fmicb.2015.01082 and references therein). Its effects on host gene expression could reflect disruption of cellular homeostasis.
RESPONSE: Thank you. Troemel lab has shown that irg-1, irg-2 and irg-3 are specific to P. aeruginosa infection and translational inhibition by the pathogen (Dunbar et al., 2012;Estes et al., 2010). We have carefully examined the literature on IPR in C. elegans. Intracellular pathogen response is induced in worms invaded by natural viruses and Microsporidia (Reddy et al., 2017;Reddy et al., 2019). We found no evidence in the literature that irg-1, irg-2, irg-3 or zip-2 transcription are components of IPR in C. elegans. They are indeed shown to be induced only in response to P. aeruginosa (Estes et al., 2010;Troemel et al., 2006). Additionally, we examined if 1-undecene exposure can induce IPR response by looking at an IPR reporter pals-5p::GFP, obtained from Troemel lab at UCSD. As shown in rebuttal Figure 10, 1undecene exposure did not induce pals-5p::GFP expression. We further confirmed that P. aeruginosa live bacteria also do not induce IPR reporter pals-5 (rebuttal Figure 10). Since pals-5 was not regulated by either P. aeruginosa or 1-undecene, we did not examine upstream regulators of pals-5 such as pals-22 or pals-25. To confirm that the strain was behaving normally, we used heat shock at 30°C and found that pals-5p::GFP expression was induced in our strain as reported by Troemel lab (Reddy et al., 2017).
Thanks for suggestion to carefully look at the effects of the bacterial volatiles, 1-undecene, DMDS etc, on plants. It has been shown in the literature that 1-undecene in combination with another volatile promotes plant growth, although molecular mechanisms remain to be deciphered. It is especially relevant because P. aeruginosa and other Pseudomonas species are associated with plants in nature. However, in fungus, Phytopthora infestans, 1-undecene has a negative impact. In the discussion section of the revised manuscript, we have discussed the effect of 1-undecene on plants and fungi. The evidence from published literature (Hunziker et al., 2015;Lo Cantore et al., 2015) and our study point to 1-undecene as a molecular pattern which mediates interkingdom interactions. These references are included in the Discussion section of the revised manuscript.
If the authors want to support their suggestion of a pathogen-specific response, they need to demonstrate that the 1-undecene-associated increases in gene expression are independent of Rebuttal Figure 10 P. aeruginosa (8h) Heat shock 30°C 24 h pals-5p::GFP pals regulation. They should also examine the expression of different markers of cellular stress. Further, they should assay the expression of candidate genes that they recently identified as being regulated by P. aeruginosa, E. faecalis and C. neoformans, such as dct-17 and lys-3.

RESPONSE:
Please see the response to the previous comment. We found that IPR reporter pals-5 was not upregulated by 1-undecene odor exposure (Rebuttal Figure 10). We also studied the activation of cytosolic stress response and oxidative stress response machinery upon exposure of worms to 1-undecene. As shown in rebuttal Figure 11, hsp-16.2p-GFP was induced by heat shock and recovery but not by 1-undecene odor exposure (Fig. S4G). Glutathione-s-transferase gst-4p::GFP was induced by exposure of worms to 20 mM paraquat but not by 1-undecene exposure (Fig. S4H in the revised manuscript and rebuttal Figure 12). Based on these experiments, we inferred that 1-undecene volatile does not cause disruption of cellular homeostasis.

Rebuttal
Other minor concerns that should be addressed: Figure 1F: N2 appear to have 2 distinct populations, one responsive and one un-responsive.
There is a similar separation in Figure S2D. What is the explanation?

RESPONSE:
We and others have found such responses in behavioral assays. This may be linked to the expression of receptors for specific stimuli or epigenetic modification. We currently have no data to provide a solid explanation for this.
Welch's t-test is insensitive to equality of the variances, but does assume normal distribution. It should not be used for non-normal data (like Figure 1F).

RESPONSE:
Thanks for your suggestion. We have carefully looked at our data and most of the data follow normal distribution. Therefore, we have applied two-tailed, unpaired t-test for analyses of choice index data presented in Figure 1F in the revised manuscript. This has been used by many other groups for the analysis of choice indices (Harris et al., 2014;Pereira et al., 2020;Worthy et al., 2018).
If the control data for Figure 4C is identical to the data in Figure 4A, this needs to be indicated.
RESPONSE: Thank you. We have provided separate controls across panels 4A, B, D and E for qRT-PCR in the revised Figure 4.
qRT-PCR is presented, for example, as "Relative mRNA expression N2 (1-Undecene / Naive)", but there are negative values, so this cannot be right. What is actually being represented?

RESPONSE:
The negative values are arrived at by representing FC value less than 1 as (-1/FC). This allows for better visualization of downregulation. For example, a fold change of 0.2 is a 5fold downregulation but hard to visualize on positive axis for fold change. Thus, we use this method of representation (Dasgupta et al., 2020;Dixit et al., 2020).
The authors make a distinction between "younger" and "older" lawns. If 1-undecene is not produced by P. aeruginosa in liquid culture and/or at 37C, the differences observed would simply be a question of the time that the bacterial cultures have been in a situation compatible with 1-undecene synthesis. This could readily checked by assaying undA expression under the different culture conditions. RESPONSE: 1-Undecene is also produced in liquid culture (Rui et al., 2014;Timm et al., 2018). We have examined undA expression only on solid lawn where we study C. elegans-bacteria interaction and show behavioral changes. One could study the change in expression in liquid culture and change of media but that would not be relevant to our study. We also do not see an alteration in undA expression level in rhlR mutant (rebuttal Figure 8) suggesting that it may be independent of quorum sensing and population density.
As you can see, the referees appreciate the introduced revisions are overall supportive of the manuscript. The referees have a few more points that should be fairly straightforward to address. Regarding the point raised by referee #3 that the data doesn't sufficiently support that 1-undecene triggers a P. aeruginosa-specific response: the suggested experiment should be fairly straightforward to do or perhaps you have already done this? Let's discuss this further via video call or email.
When you submit your revised version will you also take care of the following points: The movie legends should be removed from the appendix and ZIPed together with each movie file. The names and callouts in text needs to be corrected to 'Movie EV#'.
Methods needs correcting to Materials and Methods.
In this revised manuscript, the authors more convincingly show that 1-undecene serves as a volatile Pseudomonas-specific pathogen associated molecular pattern. This is a significant finding of general interest. In my opinion, the authors have thoroughly addressed my concerns and I now consider this manuscript suitable for publication.
There do remain some minor proofreading or textual issues, some of which are listed below: Calcium imaging figures do not indicate n or the number of imaging sessions. Figure 3E -does not indicate under what imaging condition this representative image was taken. Figure 4F -should indicate whether worms were exposed on partial or full lawns. The statement (line 219), "We did not observe calcium response to 1-undecene in AWA, AWCon or AWCoff neurons (Fig. S3A) -There appear to be small responses in AWC, so would it not perhaps be more accurate to state that you did not observe large responses in AWC?
Referee #2: The authors are responsive to my comments and addressed the concerns in the revised manuscript.
Referee #3: The manuscript is much improved. I do still, however, have reservations concerning one of the authors' central claims, exemplified by the sentence in the Discussion, "The most relevant evidence for 1-undecene as a pathogen-associated molecular pattern comes from the fact that exposure of worms to volatile alone induces upregulation of immune response genes specific to P. aeruginosa".
Yet, as I pointed out, irg-2 expression is induced by Orsay virus, and irg-1 and irg-2 expression is strongly induced by cycloheximide. Therefore, their induction is not "specific to P. aeruginosa infection and translational inhibition by the pathogen", but to translational inhibition more generally. I would thus have expected the authors to have looked, for example, at atf-4 expression after 1undecene exposure (by qRT-PCR).
These reservations are also due to the selective manner in which the authors interpret their data. They report a 2-fold increase for irg-1 expression as an induction (Fig. 4A), yet while the expression of both fmo-2 and acs-2 increased by more than 2-fold wrote, "We found that 1-undecene odor exposure did not induce expression of transcripts of fmo-2, acs-2, lipl-1, lipl-3, cpr-4, cpr-5, asp-14 and lys-3 (Fig. S4D)". They can't have it both ways.
Further they write, "We also analyzed the induction of heat shock response or oxidative stress response using hsp-16.2p::GFP and gst-4p::GFP respectively (Link et al., 1999;Link and Johnson, 2002). We found that 1-undecene odor exposure did not disrupt cellular homeostasis". These results (for which the reader should be referred to FigS4G,H) are not entirely convincing, in part as the positive controls give such weak signals. Their own results for fmo-2 show how qRT-PCR is a more sensitive assay for changes in gene expression than a GFP reporter. Is there a reason why a more sensitive, qRT-PCR analysis was not performed?
As a very minimum, the authors need to tone down substantially their claim that 1-undecene is triggering a P. aeruginosa-specific response, but ideally conduct the experiments to define whether or not 1-undecene has more general effects on host physiology, by measuring for example, atf-4 mRNA levels before and after exposure.
Minor points "qRT-PCR is presented, for example, as "Relative mRNA expression N2 (1-Undecene / Naive)", but there are negative values, so this cannot be right. What is actually being represented? RESPONSE: The negative values are arrived at by representing FC value less than 1 as (-1/FC). This allows for better visualization of downregulation. For example, a fold change of 0.2 is a 5fold downregulation but hard to visualize on positive axis for fold change".
That's fine, but this information must be included in the figure legends! When I wrote, "There are some recent studies that should have been cited, including Ringstad's on O2/CO2 sensing", I was referring to "Toll-like Receptor Signaling Promotes Development and Function of Sensory Neurons Required for a C. elegans Pathogen-Avoidance Behavior". Brandt JP, Ringstad N. Curr Biol. 2015. This needs to be included at, "except TOL-1 which has a limited role in C. elegans microbe interactions (Pradel et al., 2007;Tenor and Aballay, 2008)".

Referee #1:
In this revised manuscript, the authors more convincingly show that 1-undecene serves as a volatile Pseudomonas-specific pathogen associated molecular pattern. This is a significant finding of general interest. In my opinion, the authors have thoroughly addressed my concerns and I now consider this manuscript suitable for publication.

RESPONSE: Thank you.
There do remain some minor proofreading or textual issues, some of which are listed below: Calcium imaging figures do not indicate n or the number of imaging sessions.

RESPONSE:
Number of animals imaged in a single session are included the legend for Figure 3. Two to three imaging sessions were performed for each strain. RESPONSE: 3E represents GCaMP3 fluorescence in one of the AWB neurons 129 s after 1-undecene exposure and after 1.2 seconds (131.2 seconds on x-axis) of 1-undecene withdrawal. This information is available in the legend of Figure 3. Figure 4F -should indicate whether worms were exposed on partial or full lawns.

RESPONSE:
The worms were exposed on full lawns of P. aeruginosa.
The statement (line 219), "We did not observe calcium response to 1-undecene in AWA, AWCon or AWCoff neurons (Fig. S3A) -There appear to be small responses in AWC, so would it not perhaps be more accurate to state that you did not observe large responses in AWC?

RESPONSE:
We have modified our statement to 'We observed none or very small calcium response to 1-undecene withdrawal in AWA, AWC on or AWC off neurons'.

Referee #2:
The authors are responsive to my comments and addressed the concerns in the revised manuscript. RESPONSE: Thank you.

Referee #3:
The manuscript is much improved. I do still, however, have reservations concerning one of the authors' central claims, exemplified by the sentence in the Discussion, "The most relevant evidence for 1-undecene as a pathogen-associated molecular pattern comes from the fact that exposure of worms to volatile alone induces upregulation of immune response genes specific to P. aeruginosa".
Yet, as I pointed out, irg-2 expression is induced by Orsay virus, and irg-1 and irg-2 expression is strongly induced by cycloheximide. Therefore, their induction is not "specific to P. aeruginosa infection and translational inhibition by the pathogen", but to translational inhibition more generally. I would thus have expected the authors to have looked, for example, at atf-4 expression after 1-undecene exposure (by qRT-PCR).

RESPONSE:
Thanks for your suggestions. We have performed atf-4 (Glover-Cutter et al., 2013) qRT-PCR in worms exposed to 1-undecene odor. As shown in the rebuttal figure 1, we see just about 1.3-fold increase in atf-4 transcript in 1-undecene exposed worms compared to the naïve worms. We also examined an endoplasmic reticulum chaperone, hsp-4, (Shen et al., 2001) and found no significant change due to 1-undecene exposure. We have included this data in Appendix Figure S4I in the revised manuscript.

Rebuttal Figure 1:
Real time PCR analysis of transcripts for atf-4 and hsp-4 in N2 worms exposed to 1-undecene odor for 2 hours over naive N2 worms. n = 3. Error bars indicate SEM.

Rebuttal Figure 1
These reservations are also due to the selective manner in which the authors interpret their data. They report a 2-fold increase for irg-1 expression as an induction (Fig. 4A), yet while the expression of both fmo-2 and acs-2 increased by more than 2-fold wrote, "We found that 1-undecene odor exposure did not induce expression of transcripts of fmo-2, acs-2, lipl-1, lipl-3, cpr-4, cpr-5, asp-14 and lys-3 (Fig. S4D)". They can't have it both ways.

RESPONSE:
Thanks for your comments. We have modified our statement in the revised manuscript to 'We found that 1-undecene odor exposure showed little or no induction in expression of transcripts of fmo-2, acs-2, lipl-1, lipl-3, cpr-4, cpr-5, asp-14 and lys-3'. Both fmo-2 and acs-2 are induced tens to hundreds of folds during infection of worms with E. faecalis or pathogenic yeast Cryptococcus neoformans (Dasgupta et al., 2020) while we see 2-3-fold increase in 1-undecene exposed worms. The 2-fold change in exposed over naive worms is rather weak in comparison to induction see with pathogenic Gram-positive bacteria or yeast.
Further they write, "We also analyzed the induction of heat shock response or oxidative stress response using hsp-16.2p::GFP and gst-4p::GFP respectively (Link et al., 1999;Link and Johnson, 2002). We found that 1-undecene odor exposure did not disrupt cellular homeostasis". These results (for which the reader should be referred to FigS4G, H) are not entirely convincing, in part as the positive controls give such weak signals. Their own results for fmo-2 show how qRT-PCR is a more sensitive assay for changes in gene expression than a GFP reporter. Is there a reason why a more sensitive, qRT-PCR analysis was not performed? RESPONSE: Thank you for your suggestions. We have analyzed transcript levels of hsp16.2 and gst-4 by qRT-PCR. As shown in the rebuttal figure 2, we see no significant change in the transcript levels for hsp16.2 and two-fold increase for gst-4.

Rebuttal Figure 2:
Real time PCR analysis of transcripts for hsp-16.2 and gst-4 in N2 worms exposed to 1-undecene for 2 hours over naive N2 worms. n = 3. * P ≤ 0.05 as determined by two-tailed unpaired t-test with Welch's correction.
Error bars indicate SEM. As a very minimum, the authors need to tone down substantially their claim that 1undecene is triggering a P. aeruginosa-specific response, but ideally conduct the experiments to define whether or not 1-undecene has more general effects on host physiology, by measuring for example, atf-4 mRNA levels before and after exposure.
RESPONSE: qRT-PCR analyses of 8 immune effectors specific to pathogenic yeast and Gram-positive bacteria (Appendix Fig. S4D) indicated that they are either not induced at all or induced to about two folds for fmo-2 and acs-2. This is miniscule in comparison to 100fold or higher induction seen for fmo-2 in response to infection with pathogenic bacteria (Dasgupta 2020). In addition, we find less than two-fold change in levels of transcripts for atf-4, hsp-4 and hsp16.2, and small increase in gst-4 transcript (Rebuttal Figs. 1 and 2). Microscopy reveals no increase in reporter expression for either hsp16.2 or gst-4 ( Fig. S4G  and S4H).

Minor points
"qRT-PCR is presented, for example, as "Relative mRNA expression N2 (1-Undecene / Naive)", but there are negative values, so this cannot be right. What is actually being represented? RESPONSE: The negative values are arrived at by representing FC value less than 1 as (-1/FC). This allows for better visualization of downregulation. For example, a fold change of 0.2 is a 5-fold downregulation but hard to visualize on positive axis for fold change". That's fine, but this information must be included in the figure legends!