Heat-off responses of epidermal cells sensitize Drosophila larvae to noxious inputs

  1. Jiro Yoshino
  2. Avery Chiu
  3. Takeshi Morita
  4. Chang Yin
  5. Federico M. Tenedini
  6. Takaaki Sokabe
  7. Kazuo Emoto
  8. Jay Z. Parrish

Reviewed by

Read the full article See related articles

Discuss this preprint

Start a discussion What are Sciety discussions?

Listed in

Log in to save this article

Abstract

Perception of external thermal stimuli is critical to animal survival, and although an animal’s skin is the largest contact surface for thermal inputs, contributions of skin cells to noxious temperature sensing have not been extensively explored. Here, we show that exposure to heat transiently sensitizes Drosophila larvae to subsequent noxious stimuli. This sensitization is induced by prior stimulation of epidermal cells but not nociceptors, suggesting that epidermal cells modulate nociceptor function in response to heat exposure. Indeed, we found that Drosophila epidermal cells are intrinsically thermosensitive, exhibiting robust heat-off responses following warming to noxious temperatures as well as responses to cooling below comfortable temperatures. Further, we found that epidermal heat-off calcium responses involve influx of extracellular calcium and require the store-operated calcium channel Orai and its activator Stim. Finally, epidermal heat-off responses and heat-evoked nociceptive sensitization exhibit similar temperature dependencies, and we found that Stim and Orai are required in epidermal cells for heat-evoked nociceptive sensitization. Hence, epidermal thermosensory responses provide a form of adaptive sensitization to facilitate noxious heat avoidance.

Article activity feed

  1. Review Commons

    Note: This response was posted by the corresponding author to Review Commons. The content has not been altered except for formatting.

    Learn more at Review Commons

    Reply to the reviewers

    Detailed point-by-point response

    __ __The Reviewers provided suggestions to improve the manuscript, most notably by adding experiments to (1) further support the role of Stim and Orai in epidermal heat-off responses and (2) further characterize the thermosensory responses of epidermal cells. We additionally propose to include a new set of calcium imaging experiments to visualize nociceptor sensitization by epidermal cells.

    Reviewer #1 (Evidence, reproducibility and clarity (Required)):

    Summary:* Drosophila larvae are known to respond to noxious stimuli by rolling. The authors propose that this response arises not only by sensory response of nociceptive neurons but also by direct response of larval epidermal cells. They go onto test this idea by independently manipulating epidermal cells and nociceptive sensory neurons using GAL4 lines, GCAMPs and RNAis. The behavioural data are convincing and presented clearly with good statistical analysis. However the involvement of epidermal cells in evoking the behaviour as well as STIM/Orai mediated Ca2+ entry requires further experiments. Use of another independent GAL4 strain for epidermal cells, alternate RNAi lines for STIM and Orai, mutants for STIM and Orai and overexpression constructs for STIM and Orai would significantly enhance the data. Thus, as of now the key results require more convincing. The following additional experiments would be required to support their claims:*

    1) Either use a second epidermal GAL4 strain to show key results OR provide images of the epidermal GAL4 expression double-labelled with a ppk driver using a different fluorescent protein to establish NO overlap of the epidermal GAL4 with neurons. These strains should be available free in Bloomington.

    We agree that the specificity of the GAL4 driver is an important point. In a recent publication (Yoshino et al, eLife, 2025) we provide the most comprehensive analysis of larval epidermal GAL4 drivers published to date. Included in this study is expression analysis of R38F11-GAL4 demonstrating that it is indeed specifically expressed in the epidermis. Based on the detailed expression analysis and functional analysis provided in that paper, R38F11-GAL4 was chosen for these studies as it is both highly specific for epidermal cells and provides uniform expression across the body wall.

    In our revised manuscript, we will more clearly detail how the driver was chosen for this study and provide a citation to the prior work to accompany our description of R38F11-GAL4 as an epidermis-specific driver line.

    1. Authors need to provide better data for the involvement of STIM and Orai in the Calcium responses observed. A single RNAi for each gene with marginal change in response is insufficient. The authors also do not state if the RNAis used are validated by them or anyone else. Minimally they should repeat their experiments with at least one other validated RNAi and rescue these with overexpression constructs of STIM and Orai (available in Bloomington). It is well established in literature that overexpression of STIM/Orai can rescue SOCE in Drosophila. Ideally, to be fully convincing they should test a Drosophila knockout for STIM (available in Bloomington). Heterozygotes of this are viable and should be tested. Additionally a UAS Orai dominant negative (OraiDN) strain is available in Bloomington and can be tested.

    We appreciate the Reviewer’s perspective on the importance of characterizing the efficacy of the reagents we used in this study. However, we disagree with the characterization of the change in response as “marginal”. Our results demonstrate that epidermal knockdown of Stim or Orai causes a significant reduction in the heat-off response of epidermal cells and heat-induced nociceptive sensitization.

    In a prior published study (Yoshino et al, eLife, 2025) we validated for their efficacy of these RNAi lines in combination with the same GAL4 driver at the same developmental stage. Specifically, we demonstrated that R38F11GAL4-mediated expression of UAS-Stim RNAi or UAS-Orai RNAi significantly attenuated store operated calcium entry following story depletion by thapsigargin. In the revised manuscript, we will add a statement referring to this prior validation along with a citation. In light of this prior characterization, we disagree that additional RNAi lines are required to corroborate the results.

    The most salient point of the Reviewer’s comment is that additional evidence should be provided to demonstrate more convincingly the requirement of Stim/Orai in epidermal heat-off responses. We detail our plans to address this point below, but first address the specific experimental suggestions the Reviewer provides.

    First, the Reviewer suggests the use of a dominant-negative version of Orai, and we agree that this could prove complimentary to our RNAi experiments.

    The Reviewer suggests two additional genetic approaches which are well-reasoned but problematic. First, they suggest rescuing the RNAi knockdowns with overexpression approaches. In addition to requiring the generation of new, RNAi-refractory transgenes, this approach is confounded by the effects of overexpressing CRAC channel components. Orai channels exhibit highly cooperative activation by Stim, and we previously showed that epidermal Stim overexpression drove mechanical nociceptive sensitization. Although this dosage effect confounds the rescue assays, we will examine whether epidermal Stim overexpression similarly sensitizes larvae to noxious thermal inputs as we would predict from our model.

    The final experiment the Reviewer suggests – phenotypic analysis of Stim knockouts – is not possible due to the lethal phase of the mutants. Furthermore, it is not possible using traditional mosaic analysis to generate mutant epidermal clones that span the entire epidermis. Such an approach might be possible with a newly engineered FLP-out Stim allele, but generating that reagent is beyond the scope of this work. The Reviewer suggests characterization of Stim heterozygotes, but Drosophila genes rarely show strong dosage effects as heterozygotes (though we acknowledge that dosage effects can be amplified in the cases of genetic interactions), hence a negative result (no effect on heat-off responses) would not be meaningful. In principle we could test whether Stim hetorozygosity enhances effects of epidermal Stim RNAi. Although a negative result will not be telling, the experiment is straightforward, and an enhancement of the effect of Stim RNA would support the model that RNAi provides an incomplete functional knockdown of Stim. We will therefore perform this experiment and incorporate the results into the revised manuscript, pending a postitive outcome.

    To better define the contributions of Stim and Orai to heat-off responses of epidermal cells, we will incorporate results from the following new experiments into our revised manuscript:

    • We will monitor effects of epidermis-specific expression of a dominant negative form of Orai on epidermal heat-off responses (calcium imaging) and heat-induced nociceptive sensitization (behavioral assays).
    • We will monitor effects of epidermis-specific co-expression of Stim+Orai RNAi on epidermal heat-off responses (calcium imaging) and heat-induced nociceptive sensitization (behavioral assays)
    • Orai channels exhibit highly cooperative activation by Stim, therefore we will examine whether epidermal Stim overexpression increases the amplitude of heat-off responses (calcium imaging) and sensitizes larvae to noxious thermal inputs (behavioral assays) as we would predict from our model.

    Minor comments that can be addressed:

    1. Figure 1: Further details required on how the rolling response is measured. Figure is uninformative. A video would be really helpful.

    We appreciate the suggestion. We will add a more detailed explanation of how the behaviors were scored along with an annotated video.

    1. I could not find Figure 1I described in the text. This section should be explained properly.

    Figure 1I is described in the figure legend and we will add an in-text citation.

    1. Figure 3: There appears to a small response at 32oC - why is this ignored in the text? It would be useful to have S3 in the main figure.

    The small response at 32C is not ignored, though that individual response is better understood in the context of all responses plotted in Figure 3D. We will reword the phrase “At temperature maxima below 35°C epidermal cells rarely exhibited heat-off responses” to reflect the small response that is observed at lower temperatures. We will also replace the trace in the figure – the original submission contained the one outlier sample that exhibited robust responses at 32 C.

    We appreciate the suggestion to include Fig S3 in the main text – we initially included it, but moved it to the supplement for space considerations. We will include it as a main figure in our revised submission.

    1. Fig 4: The DF/F traces for the two RNAis should be included in this figure.

    We appreciate the suggestion; we will add these traces to our revised submission.

    1. Extent of knockdown in the epidermis by each RNAi should be shown by RTPCRs.

    We note that efficacy of the knockdowns has been validated by us in acutely dissociated epidermal cells. RTPCR validation as described would require FACS-sorting of acutely dissociated, GFP-labeled epidermal cells from each specimen, an extremely time- and resource intensive experiment that provides limited information. The more relevant information is the physiological readout of Stim/Orai functional knockout using these reagents which we previously conducted. As described above, we will add a description of these experiments and the relevant citation.

    1. The authors need to explain why only a small change in the Ca2+ response is seen with either RNAi. Are there other Ca2+ channels involved? Ideally they could test mutants/RNAi for the TRP channel family. Loss of SOCE in Drosophila neurons changes the expression of other membrane channels - is this possible here? Minimally, this possibility needs to be discussed.

    We agree with the Reviewer that this topic warrants further discussion. Pending the results of our planned experiments (Orai dominan negative, Stim+Orai RNAi), we will incorporate a discussion of other channels that may contribute to the heat-off response. We appreciate the Reviewers point that loss of SOCE in Drosophila neurons can change the expression of membrane channels – that is an intriguing possibility that might explain the modest effects of Stim or Orai knockdown. We have not investigated effects of epidermal Stim/Orai knockdown on expression of other channels, but will incorporate this possibility into our discussion.

    1. In the methods section please explain how the % DF/F calculations are done and how are they normalised to the ionomycin response.

    We will incorporate these additional details in the methods section.

    1. Authors need to look at previous work on STIM and Orai in Drosophila and reference appropriately.

    We appreciate the suggestion and will incorporate additional discussion of relevant Drosophila work on STIM and Orai.

    **Referees cross-commenting**

    Reviewers 2 and 3 have raised some additional queries to what I had mentioned in my review. I agree with their comments. The authors should attempt to address all comments by all three reviewers.

    We address their comments below.

    Reviewer #1 (Significance (Required)):

    This is an interesting study that identifies epidermal cells in Drosophila with the ability to sense a drop in temperature after receiving noxious heat stimuli and invoke appropriate behaviour. Behaviour experiments are well conducted and convincing. So far only nociceptive neurons were thought to control such behavioural responses so the work is significant and important for the field. The mechanism identified needs further convincing and I have suggested experiments that would be of help. With the additional experiments suggested the work will be of interest to neuroethologists, Drosophila neuroscientists and scientists in the field of Ca signaling.

    Reviewer #2 (Evidence, reproducibility and clarity (Required)):

    Summary:

    Noxious heat can have a strong adverse effect on animals, resulting in sensitization when noxious thermal stimuli are applied repeatedly. Noxious heat induces a characteristic rolling behavior in Drosophila melanogaster larvae. This study investigates sensitization, whereby a second heat stimulus evokes this behavior with significantly shorter latency (e.g., 3.4 seconds) than the initial exposure (e.g., 8.79 seconds). While prior research has implicated central and peripheral neurons in this process, recent findings in mammalian systems suggest a role for keratinocytes.

    In this manuscript, Yoshino et al. report that epidermal cells are necessary and sufficient to mediate heat sensitization in D. melanogaster larvae. Using an ex vivo epidermal imaging system, the authors demonstrate that calcium influx in epidermal cells is crucial for sensitization. Importantly, this calcium influx was observed only when the temperature was lowered from a dangerously high to a safe temperature. The calcium channel system Orai and Stim facilitates this influx.

    Major comments:

    (1) The authors clearly demonstrate the heat-off reaction using calcium influx imaging. However, all of the imaging shows the response to the first stimulation. Since the study focuses on sensitization, which shows a quicker response to the second heat stimulus, it would be helpful if the authors showed calcium influx when the second stimulus was applied. It would also be interesting to see how many times the epidermal cells can react to heat stimulation.

    We appreciate the suggestion from the Reviewer but note that the calcium influx we show occurs in epidermal cells, which signal to neurons to potentiate future responses in our model. We have emphasized this point in our revised manuscript.

    The relevant response to visualize the sensitization is the heat-evoked calcium response in nociceptors, not epidermal cells. We have verified that C4da neurons exhibit calcium responses to the warming stimulus we use in our heat-off paradigm and our preliminary studies suggest that the heat-off stimulus potentiates future responses to noxious heat in nociceptors. We will therefore examine (1) whether epidermal stimulation triggers a sensitization of nociceptors to thermal stimuli by monitoring heat-induced calcium responses using GCaMP, and (2) whether epidermal Stim and Orai are required for this sensitization.

    The second comment addresses the response of epidermal cells to repeated rounds of stimuli. We agree that this is an interesting point. We have verified that epidermal cells indeed respond to multiple rounds of heat-off stimuli. We will incorporate results from a paradigm in which epidermal cells are presented with two successive heat-off stimuli, spaced by 5 minutes to allow epidermal cytosolic calcium to return to baseline. We will incorporate new analysis examining the relative magnitude of epidermal cells to the first and second stimulus.

    (2) Figure 5 only shows one condition: a 30-second interval between the first and second heat application. While the rolling latency of the Luciferase RNAi control ranges from 4 to 12 seconds (with a median of 5 seconds), Fig. 1E shows a latency ranging from 6 to 12 seconds (with a median of 10 seconds) under the same 30-second interval conditions. This difference makes interpreting the effect of Stim and Orai confusing. The authors need to clarify whether the knockdowns accelerate the first response or delay the second response.

    The Reviewer notes that we assayed effects of Stim/Orai RNAi on heat-induced nociceptive sensitization in only one paradigm. Given the kinetics of cytosolic calcium increases following Stim or Orai RNAi in epidermal cells (Fig. 4F), we agree that an additional set of behavior experiments investing sensitization following a 60 sec recovery is warranted. For our revision we will conduct a time-course to assay requirements of epidermal Stim and Orai (using epidermal expression of Stim/Orai RNAi and Orai dominant negative transgenes) on heat-induced nociceptive sensitization. Our preliminary studies indicate that Stim and Orai RNAi significantly reduce heat-induced sensitization following 60 s of recovery (we present results from 30 s of recovery in the original submission).

    The Reviewer raises some questions about differences in behavioral latencies in Figure 1E and Figure 5B. We intentionally avoid such comparisons both because the genetic backgrounds are different and the experiments were conducted at very different times (more than 1 year apart). In both experiments the salient feature that we discuss is the presence or absence of sensitization, not the mean latency. We note that we do compare mean latency values in Figure 1B, but that was a distinct experimental paradigm (global heat of variable temperatures followed by focal noxious heat) designed specifically to define heat stimuli that generate the maximum level of sensitization. In that case, the genotype was fixed and all assays were conducted concurrently.

    Minor comments:

    (i) In Fig. 2C´´, the authors observed clear calcium influx in epidermal cells by combining the GCaMP genetic tool with an ex vivo thermal perfusion system. Although this system applies heat uniformly across the epidermal tissue, calcium influx is spatially restricted, appearing primarily in the head and tail regions of the epidermis. These results suggest that the heat-responsive epidermal cells are localized to these regions or that there are regional differences in sensitivity. The authors should explain the spatial relationship between the heat-applied epidermal cells and the occurrence of calcium influx.

    The Reviewer notes that intensity of the epidermal GCaMP signal is particularly intense in the anterior and posterior portions of the fillet preparation (Fig. 1B-1C), and we agree that it would be useful to include an explanation of this result, which is an artifact of the sample preparation.

    The specimens we use for calcium preparation are “butterfly” preparations – the body wall is filleted along the long axis with the exception of regions at the head and tail that are pinned down on sylgard plates. Hence, the regions in the head and tail contain intact tissue (including a double layer of skin when we image in widefield), not a single layer of skin (the rest of the prep). More significantly, the head and tail regions are pinned down, creating a wound that triggers lasting local calcium transients (note signal in the absence of temperature stimulus, Figure 1B’ and 1B”, 1C’). We therefore exclude this region from our analysis. We note that our behavior studies relied on stimuli presented to the abdominal segments we sample in the semi-intact calcium imaging. Similarly, we dissociated epidermal cells exclusively from these segments for imaging of acutely isolated epidermal cells.

    We do note that there is a periodicity to the signal – within each segment there are local maxima and minima of signal, and we agree with the Reviewer that this spatial segregation is an interesting point for discussion. We will add 1-2 sentences to our discussion of the result to acknowledge this point.

    (ii) Related to comment (i) above, if heat stimuli are applied topically using a heat probe under the ex vivo imaging system, how large an area reacts to the stimuli?

    The Reviewer raises an interesting question about the local response to heat stimuli. In our dissociated cell experiments we found that the overwhelming majority of isolated epidermal cells exhibit heat-off responses, and we likewise find that the majority of cells in our semi-intact preparation respond to heat-off stimuli. However, our current probe for delivering local heat stimuli is not compatible with our imaging system. We are working to incorporate an IR laser to focally deliver heat stimulus to explore whether epidermal cells signal to neighbors following stimulation, but such studies are beyond the scope of the current work.

    (iii) Providing supplementary movie(s) of the calcium live imaging would enhance the reader's understanding.

    We agree with the Reviewer that this would be a useful supplement. We will add representative movies as experimental supplements in our revised manuscript.

    (iv) The time point of the image in Fig. 2C´ ("before heat") is not the most informative for demonstrating a "heat-off" response. The authors should replace it with an image taken during the heat application to provide a more direct comparison with the post-stimulus influx shown in Fig. 2C´´.

    We appreciate the Reviewer’s suggestion and agree this would be a better choice to visually represent the change in fluorescence induced by the heat-off response. We will make this change in our revised manuscript.

    (v) The authors state that sensitization occurs "primarily in the 30-45 ºC range." However, the rolling probability and latency developed oppositely at 45 ºC stimulation than at 40 ºC. It would be doubtful that 45 ºC may be approaching a noxious or damaging threshold that engages a different phenomenon. The authors should reconsider including 45 ºC within the optimal sensitization range or provide a justification.

    We agree with the Reviewer that a more detailed discussion of the effects of temperature at the end of the range (45 C) is warranted. Exposure to a 45 C global heat stimulus triggered temporary paralysis in some larvae, and we suspect that this accounts for the apparent reduction in roll probability following the second stimulus. We can add a plot depicting the proportion of larvae that exhibited paralysis during 45 C global heat and determine whether these heat-paralyzed larvae exhibited distinct responses from larvae that were not paralyzed and provide a more detailed account of the optimal sensitization range.

    Treatment with 45 C stimuli still triggered a significant reduction in roll latency (sensitization), but we did not examine whether the latency was significantly different from what was observed at 40 C. We can add that analysis in the revision.

    (vi) In the sentence "To this end, we developed a perfusion system, that would deliver thermal ramps from ~20-45ºC ...," the tilde ~ should be replaced with "approximately".

    Noted. We will make the change.

    (vii) Throughout the manuscript, please clarify in the figure legends whether the sample size (n) refers to the number of individual animals or the number of cells.

    Noted. We will add the relevant details to our sample sizes notations.

    (viii) The Key Resources Table does not specify the wild-type (WT) strain used for the control experiments (e.g., in Fig. 1). Please provide the full genotype of the control strain used.

    We included the experimental genotypes in each figure legend, which we find more useful than the key resource table, which contains a list of all reagents used in the study (Drosophila alleles included).

    Reviewer #2 (Significance (Required)):

    General Assessment

    This study addresses a fundamental question in sensory biology: whether epidermal cells, long regarded as passive participants in somatosensation, actively contribute to noxious heat detection and avoidance behavior. While previous work has defined the neuronal circuits and TRP channel mechanisms underlying thermal nociception in Drosophila larvae, the potential sensory role of skin cells has remained largely unexplored. The authors integrate behavioral analysis with in vitro and ex vivo calcium imaging to provide a rigorous, multi-level investigation of epidermal thermosensitivity.

    Advancement

    The work advances the field by revealing that Drosophila epidermal cells are intrinsically thermosensitive and can acutely sensitize larval nociceptive responses to noxious heat through heat-off signaling. This discovery shifts the current paradigm of thermal nociception from a neuron-centric model to one that incorporates epidermal contributions, highlighting a conserved and previously underappreciated role of skin cells in active environmental sensing.

    The reviewer's expertise: Molecular genetics, developmental biology, insect physiology and endocrinology.

    Reviewer #3 (Evidence, reproducibility and clarity (Required)):

    This manuscript describes the temperature responses of Drosophila larval epidermal cells. These cells are activated by cooling and also exhibit strong heat-off responses. Orai and Stim are required in epidermal cells for these heat-off responses. The heat-off responses sensitize the epidermal cells, leading to a greater proportion of animals displaying rolling behaviors and a reduced latency to initiate rolling following noxious heating treatment. The following comments are intended to help improve the manuscript.

    Major:

    1. In Figure 3A, the conclusion will be strengthened by testing heat-off responses from 10 {degree sign}C to 40 {degree sign}C.

    The Reviewer makes an important point. In our original experiment, the lack of response in the 10C – 30C experiment could be due to some cold-induced suppression of the off response. We have found that this is not the case – we have found that off responses following a 10C-40C ramp are indistinguishable from responses to a 20C-40C ramp. In our revised manuscript we will incorporate new results showing epidermal heat off responses to a 10C-40C ramp as well as normalization to 20C-40C responses performed in parallel.

    Figure 4C shows that 2-APB suppresses the heat-off response. Since 2-APB blocks both Orai and TRP channels, it is unclear why the authors focused exclusively on the Orai pathway without testing TRP channels.

    We found that epidermal cells exhibited minimal responses to warming stimuli, as would be expected for the epidermally expressed TRP channel TRPA1. In addition, the heat-off response we identified was remarkably similar to characteristic heat-off responses of mammalian CRAC channels. Hence, we focused our attention on the Orai pathway. While we agree that contributions of TRP channels could be of interest, especially if our additional analyses (double RNAi and Orai Dominant Negative) support the model that additional channels likely contribute to the heat-off response, the characteristic temperature responses of CRAC channels made them the most plausible candidate.

    In parallel to the experiments to further characterize Stim/Orai contributions to the heat-off response, we will assay requirements of TRPA1 to heat-induced nociceptor sensitization.

    While 2-APB completely abolishes the heat-off response, Orai and Stim RNAi only slightly (although significantly) reduce calcium responses. The knockdown efficiency of the RNAi constructs should be validated. Furthermore, testing whether combining Orai RNAi and Stim RNAi produces a stronger reduction in calcium responses would be informative.

    We addressed the question of knockdown efficiency above, and agree that testing the effects of Orai RNAi and Stim RNAi in combination is worthwhile. We detailed our plans for these experiments above.

    The study uses third-instar larvae. Please specify whether early, mid, or late third instar were used.

    In our original submission we stated “Third-instar larvae (96-120 AEL) larvae were used in all experiments” We provide additional details on the staging of larvae for all experiments in the methods section of our revised submission. To synchronize cultures, embryos were collected from experimental crosses for 24 h, aged for 96 h, and foraging mid-third instar larvae (96-120 h old) were used for all experiments.

    Please provide more details about the thin layer of water used. Specifically, indicate the size of the Peltier plate and the volume of water applied.

    We provide additional details on the application of global heat stimulus in the methods section of our revised manuscript. “For assays testing effects of varying the temperature of prior thermal stimuli on thermal nociception, larvae were individually transferred to a pre-warmed Peltier plate (11 x 7 cm; Torrey Pines Scientific). Peltier plates were warmed to the indicated temperatures, a thin layer of water was applied to the surface using a paint brush, and the temperature was verified using an infrared thermometer. Larvae were transferred individually to the Peltier plate, incubated for the indicated time, and recovered to 2% Agar Pads using a paint brush. Following 10 s of recovery, larvae were stimulated with a 41.5°C thermal probe, as above, and latency to the first complete roll was recorded.”

    Minor:

    1. There is an inconsistency between the text and the figure regarding the sample number in Figure 1D.

    We thank the reviewer for identifying the discrepancy. This inconsistency has been corrected in the revised submission.

    Please provide the raw representative data for the time course of heat-off calcium responses in Figure 1E.

    We will incorporate representative traces for the heat-off responses plotted in Figure 1E.

    A period is missing at the end of the sentence: "For curve fitting, sample-averaged fluorescence traces were fitted with a single exponential decay function using R to extract a representative time constant (τ) and assess response kinetics."

    We thank the reviewer for identifying the omission. The period has been added.

    In the sentence "Behavior Responses were analyzed post-hoc blind to genotype and were plotted according to roll probability and roll latency," the word Responses should begin with a lowercase r.

    This has been corrected in the revised submission.

    Reviewer #3 (Significance (Required)):

    This manuscript describes the heat-off responses of larval epidermal cells and investigates their underlying molecular mechanisms as well as associated behavioral consequences.

    The calcium responses and behavioral assays are clearly presented. However, the contribution of Stim and Orai to this process is not convincing.

    The study may be of interest to researchers working on Drosophila and temperature sensation, as well as to those studying Orai and Stim function.

    I am a researcher specializing in Drosophila thermosensation.

  2. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #3

    Evidence, reproducibility and clarity

    This manuscript describes the temperature responses of Drosophila larval epidermal cells. These cells are activated by cooling and also exhibit strong heat-off responses. Orai and Stim are required in epidermal cells for these heat-off responses. The heat-off responses sensitize the epidermal cells, leading to a greater proportion of animals displaying rolling behaviors and a reduced latency to initiate rolling following noxious heating treatment. The following comments are intended to help improve the manuscript.

    Major:

    1. In Figure 3A, the conclusion will be strengthened by testing heat-off responses from 10 {degree sign}C to 40 {degree sign}C.
    2. Figure 4C shows that 2-APB suppresses the heat-off response. Since 2-APB blocks both Orai and TRP channels, it is unclear why the authors focused exclusively on the Orai pathway without testing TRP channels.
    3. While 2-APB completely abolishes the heat-off response, Orai and Stim RNAi only slightly (although significantly) reduce calcium responses. The knockdown efficiency of the RNAi constructs should be validated. Furthermore, testing whether combining Orai RNAi and Stim RNAi produces a stronger reduction in calcium responses would be informative.
    4. The study uses third-instar larvae. Please specify whether early, mid, or late third instar were used.
    5. Please provide more details about the thin layer of water used. Specifically, indicate the size of the Peltier plate and the volume of water applied.

    Minor:

    1. There is an inconsistency between the text and the figure regarding the sample number in Figure 1D.
    2. Please provide the raw representative data for the time course of heat-off calcium responses in Figure 1E.
    3. A period is missing at the end of the sentence: "For curve fitting, sample-averaged fluorescence traces were fitted with a single exponential decay function using R to extract a representative time constant (τ) and assess response kinetics."
    4. In the sentence "Behavior Responses were analyzed post-hoc blind to genotype and were plotted according to roll probability and roll latency," the word Responses should begin with a lowercase r.

    Significance

    This manuscript describes the heat-off responses of larval epidermal cells and investigates their underlying molecular mechanisms as well as associated behavioral consequences.

    The calcium responses and behavioral assays are clearly presented. However, the contribution of Stim and Orai to this process is not convincing.

    The study may be of interest to researchers working on Drosophila and temperature sensation, as well as to those studying Orai and Stim function.

    I am a researcher specializing in Drosophila thermosensation.

  3. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #2

    Evidence, reproducibility and clarity

    Summary:

    Noxious heat can have a strong adverse effect on animals, resulting in sensitization when noxious thermal stimuli are applied repeatedly. Noxious heat induces a characteristic rolling behavior in Drosophila melanogaster larvae. This study investigates sensitization, whereby a second heat stimulus evokes this behavior with significantly shorter latency (e.g., 3.4 seconds) than the initial exposure (e.g., 8.79 seconds). While prior research has implicated central and peripheral neurons in this process, recent findings in mammalian systems suggest a role for keratinocytes. In this manuscript, Yoshino et al. report that epidermal cells are necessary and sufficient to mediate heat sensitization in D. melanogaster larvae. Using an ex vivo epidermal imaging system, the authors demonstrate that calcium influx in epidermal cells is crucial for sensitization. Importantly, this calcium influx was observed only when the temperature was lowered from a dangerously high to a safe temperature. The calcium channel system Orai and Stim facilitates this influx.

    Major comments:

    (1) The authors clearly demonstrate the heat-off reaction using calcium influx imaging. However, all of the imaging shows the response to the first stimulation. Since the study focuses on sensitization, which shows a quicker response to the second heat stimulus, it would be helpful if the authors showed calcium influx when the second stimulus was applied. It would also be interesting to see how many times the epidermal cells can react to heat stimulation.

    (2) Figure 5 only shows one condition: a 30-second interval between the first and second heat application. While the rolling latency of the Luciferase RNAi control ranges from 4 to 12 seconds (with a median of 5 seconds), Fig. 1E shows a latency ranging from 6 to 12 seconds (with a median of 10 seconds) under the same 30-second interval conditions. This difference makes interpreting the effect of Stim and Orai confusing. The authors need to clarify whether the knockdowns accelerate the first response or delay the second response.

    Minor comments:

    (i) In Fig. 2C´´, the authors observed clear calcium influx in epidermal cells by combining the GCaMP genetic tool with an ex vivo thermal perfusion system. Although this system applies heat uniformly across the epidermal tissue, calcium influx is spatially restricted, appearing primarily in the head and tail regions of the epidermis. These results suggest that the heat-responsive epidermal cells are localized to these regions or that there are regional differences in sensitivity. The authors should explain the spatial relationship between the heat-applied epidermal cells and the occurrence of calcium influx.

    (ii) Related to comment (i) above, if heat stimuli are applied topically using a heat probe under the ex vivo imaging system, how large an area reacts to the stimuli?

    (iii) Providing supplementary movie(s) of the calcium live imaging would enhance the reader's understanding.

    (iv) The time point of the image in Fig. 2C´ ("before heat") is not the most informative for demonstrating a "heat-off" response. The authors should replace it with an image taken during the heat application to provide a more direct comparison with the post-stimulus influx shown in Fig. 2C´´.

    (v) The authors state that sensitization occurs "primarily in the 30-45 ºC range." However, the rolling probability and latency developed oppositely at 45 ºC stimulation than at 40 ºC. It would be doubtful that 45 ºC may be approaching a noxious or damaging threshold that engages a different phenomenon. The authors should reconsider including 45 ºC within the optimal sensitization range or provide a justification.

    (vi) In the sentence "To this end, we developed a perfusion system, that would deliver thermal ramps from ~20-45ºC ...," the tilde ~ should be replaced with "approximately".

    (vii) Throughout the manuscript, please clarify in the figure legends whether the sample size (n) refers to the number of individual animals or the number of cells.

    (viii) The Key Resources Table does not specify the wild-type (WT) strain used for the control experiments (e.g., in Fig. 1). Please provide the full genotype of the control strain used.

    Significance

    General Assessment

    This study addresses a fundamental question in sensory biology: whether epidermal cells, long regarded as passive participants in somatosensation, actively contribute to noxious heat detection and avoidance behavior. While previous work has defined the neuronal circuits and TRP channel mechanisms underlying thermal nociception in Drosophila larvae, the potential sensory role of skin cells has remained largely unexplored. The authors integrate behavioral analysis with in vitro and ex vivo calcium imaging to provide a rigorous, multi-level investigation of epidermal thermosensitivity.

    Advancement

    The work advances the field by revealing that Drosophila epidermal cells are intrinsically thermosensitive and can acutely sensitize larval nociceptive responses to noxious heat through heat-off signaling. This discovery shifts the current paradigm of thermal nociception from a neuron-centric model to one that incorporates epidermal contributions, highlighting a conserved and previously underappreciated role of skin cells in active environmental sensing.

    The reviewer's expertise: Molecular genetics, developmental biology, insect physiology and endocrinology.

  4. Review Commons

    Note: This preprint has been reviewed by subject experts for Review Commons. Content has not been altered except for formatting.

    Learn more at Review Commons

    Referee #1

    Evidence, reproducibility and clarity

    Summary: Drosophila larvae are known to respond to noxious stimuli by rolling. The authors propose that this response arises not only by sensory response of nociceptive neurons but also by direct response of larval epidermal cells. They go onto test this idea by independently manipulating epidermal cells and nociceptive sensory neurons using GAL4 lines, GCAMPs and RNAis. The behavioural data are convincing and presented clearly with good statistical analysis. However the involvement of epidermal cells in evoking the behaviour as well as STIM/Orai mediated Ca2+ entry requires further experiments. Use of another independent GAL4 strain for epidermal cells, alternate RNAi lines for STIM and Orai, mutants for STIM and Orai and overexpression constructs for STIM and Orai would significantly enhance the data. Thus, as of now the key results require more convincing. The following additional experiments would be required to support their claims:

    1. Either use a second epidermal GAL4 strain to show key results OR provide images of the epidermal GAL4 expression double-labelled with a ppk driver using a different fluorescent protein to establish NO overlap of the epidermal GAL4 with neurons. These strains should be available free in Bloomington.

    2. Authors need to provide better data for the involvement of STIM and Orai in the Calcium responses observed. A single RNAi for each gene with marginal change in response is insufficient. The authors also do not state if the RNAis used are validated by them or anyone else. Minimally they should repeat their experiments with at least one other validated RNAi and rescue these with overexpression constructs of STIM and Orai (available in Bloomington). It is well established in literature that overexpression of STIM/Orai can rescue SOCE in Drosophila. Ideally, to be fully convincing they should test a Drosophila knockout for STIM (available in Bloomington). Heterozygotes of this are viable and should be tested. Additionally a UAS Orai dominant negative (OraiDN) strain is available in Bloomington and can be tested.

    Minor comments that can be addressed:

    1. Figure 1: Further details required on how the rolling response is measured. Figure is uninformative. A video would be really helpful.

    2. I could not find Figure 1I described in the text. This section should be explained properly.

    3. Figure 3: There appears to a small response at 32oC - why is this ignored in the text? It would be useful to have S3 in the main figure.

    4. Fig 4: The DF/F traces for the two RNAis should be included in this figure.

    5. Extent of knockdown in the epidermis by each RNAi should be shown by RTPCRs.

    6. The authors need to explain why only a small change in the Ca2+ response is seen with either RNAi. Are there other Ca2+ channels involved? Ideally they could test mutants/RNAi for the TRP channel family. Loss of SOCE in Drosophila neurons changes the expression of other membrane channels - is this possible here? Minimally, this possibility needs to be discussed.

    7. In the methods section please explain how the % DF/F calculations are done and how are they normalised to the ionomycin response.

    8. Authors need to look at previous work on STIM and Orai in Drosophila and reference appropriately.

    Referees cross-commenting

    Reviewers 2 and 3 have raised some additional queries to what I had mentioned in my review. I agree with their comments. The authors should attempt to address all comments by all three reviewers.

    Significance

    This is an interesting study that identifies epidermal cells in Drosophila with the ability to sense a drop in temperature after receiving noxious heat stimuli and invoke appropriate behaviour. Behaviour experiments are well conducted and convincing. So far only nociceptive neurons were thought to control such behavioural responses so the work is significant and important for the field. The mechanism identified needs further convincing and I have suggested experiments that would be of help. With the additional experiments suggested the work will be of interest to neuroethologists, Drosophila neuroscientists and scientists in the field of Ca signaling.

  5. Version published to 10.1101/2025.05.30.656957 on bioRxiv