Cytokine control of systemic hypoxia tolerance in Drosophila

  1. Kate Ding
  2. Prajakta Bodkhe
  3. Byoungchun Lee
  4. Danielle Polan
  5. Amy Wycislik
  6. Tiffany Cheung
  7. Sophie Wu
  8. Savraj S Grewal

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

Systemic hypoxia—reduction in oxygen supply to all tissues and organs—occurs in both physiological and pathological conditions in animals. Under these conditions, organisms must adapt their physiology to ensure proper functioning and survival. While extensive research has characterized how individual cells sense and adapt to low oxygen conditions, the mechanisms that coordinate whole-body responses to systemic hypoxia remain poorly understood. In this study, we uncover an interorgan signaling response mediated by the cytokine Unpaired-3 (upd3) that is important for systemic hypoxia tolerance in Drosophila . We demonstrate that hypoxia rapidly induces upd3 expression and activates JAK/STAT signaling in both larvae and adults. Interestingly, we discovered a sex-specific requirement for this pathway, with females requiring upd3 for hypoxia survival while males do not. We also identify the intestine as a critical source of hypoxia-induced upd3 and show that gut-derived upd3 signals to the fat body and oenocytes to mediate hypoxia tolerance by promoting nitric oxide synthase expression. Furthermore, we reveal an unexpected role for the canonical hypoxia response transcription factor HIF-1α/sima as a molecular brake, preventing lethal upd3 overproduction, revealing that hypoxia survival requires precise cytokine dosage control. Our findings define a gut-to-fat/oenocyte signaling axis that coordinates systemic hypoxia adaptation, highlighting the complex interplay between classic hypoxia response pathways and cytokine signaling in maintaining organismal homeostasis during oxygen limitation. This work provides important insights into how organisms adapt to systemic hypoxia, with potential implications for understanding systemic hypoxia-related human pathologies such as respiratory disorders and sleep apnea.

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

    We thank all three reviewers for their careful and constructive engagement with our manuscript. We are encouraged by their overall positive assessment of the work. Reviewer 1 described this as "an important study" that addresses a significant gap in understanding systemic, inter-organ responses to hypoxia, and noted the potential relevance of our findings to mammalian IL-6 biology. Reviewer 2 highlighted the study as being of "high significance" and described it as "a foundation study that will be the motivation for numerous high-impact papers in the future", noting its broad relevance to understanding hypoxia in both health and disease. In the revised manuscript, we have addressed all of the reviewers' comments and critiques. This includes performing several new experiments, expanding our Discussion, and making a number of clarifications to the text, figures, and methods as detailed below.

    Reviewer #1

    __(Evidence, reproducibility and clarity (Required)): __The authors describe a role of Unpaired 3 (Upd3) in tissue communication in responses to hypoxia in Drosophila adult flies. Upd3 mRNA is strongly upregulated in hypoxia, along with well-characterized JAK/STAT downstream target genes, in both adult fly males and females, as well as in larvae. Interestingly, adult females but not males require Upd3 for 15 to 24 h survival in hypoxia, as Upd3 mutant females but not males die to a much larger proportion in these conditions. Adult females they display strong hypoxic upregulation of Upd3 in the gut, assessed by RT-PCR or through a Gal4 transcriptional reporter, mainly in epithelial enterocytes. Enterocyte-specific RNAi-mediated KD indicated that this enterocyte expression of Upd3 represents about 40% of Upd3 expression in the whole body. Enterocyte-specific KD of Upd3 in adult females significantly reduced survival in hypoxia, suggesting that this expression is critical for hypoxic adaptation. Tissue-specific analysis of the expression of the STAT target genes, SOCS36E, TotA and TotM revealed that stimulation of the JAK/STAT pathway in hypoxia is widespread, although more pronounced in abdominal tissues. Indeed, overexpression of Upd3 in enterocytes provokes upregulation o both target genes TotA and TotM. Consistent with this RNAi-dependent inhibition of the JAK/STAT pathway in the fat body and oenocytes significantly reduced survival of female flies in hypoxia. Nitric oxide synthase (NOS) is strongly upregulated in adult female abdomens upon hypoxic exposure, and KD of NOS in fat body and oenocytes reduced hypoxic survival. Surprisingly, the found that ubiquitous KD of HIFa/Sima led to mitigation of Upd3 hypoxic induction and, more clearly, to JAK/STAT target gene induction. HIF KD flies displayed increased lethality in hypoxia, and this lethality was slightly mitigated in Upd3 heterozygous flies. The authors conclude that increased lethality of HIF-minus flies in hypoxia stems at least in part from excessive levels of Upd3. The authors then find that HIF/Sima-dependent inhibition of Upd3 expression is non-cell autonomous, since KD of Sima specifically in the gut does not affect expression of Upd3 in this organ. Instead, Sima KD at the fat body led to significant increase of Upd expression in the gut, suggesting that a Sima-born signal communicates these two organs, leading to restriction of Upd3 intestinal expression. ROS does not seem to be the signal that communicates the fat body with the gut, as expression of catalase in the fat body did not affect expression of Upd3 in the gut.

    (Significance (Required)): This is an important study, because most previous studies have focused on cell-autonomous responses to hypoxia, but much less is known about systemic responses to low oxygen conditions, particularly in relation to inter-organ communication during this responses. This work defines the cytokine unpaired 3, homolog of human interleukin 6, as a major regulator of systemic responses to hypoxia. Future studies will determine if interleukin 6 plays similar roles in mammals. This work might be of interest for a broad audience interested in responses to hypoxia, as well as general physiology.

    We thank Reviewer 1 for their careful reading and comments on the manuscript. We are pleased that they found this to be "an important study" that addresses a gap in understanding systemic, inter-organ responses to hypoxia. We have addressed each of their concerns in the revised manuscript as outlined below.

    __MAJOR CONCERNS __

    1. Figure 1 lacks statistical analysis. It is important to determine if the apparent differences in gene expression are statistically significant.

    We have now added the statistical analyses to the revised version of the figures.

    1. Is NOS expression in fat body/oenocytes JAK/STAT-dependent? Block the pathway in hypoxia specifically in this cells and check.

    To address this, we blocked JAK/STAT signaling specifically in fat body/oenocytes under hypoxia and examined the expression of Nos, as well as bnl and Hipk - two additional genes we find are regulated by gut-derived Upd3 and required for hypoxia tolerance.

    Interestingly, fat body/oenocyte-specific knockdown of STAT92E suppressed hypoxia-induced Hipk expression but did not affect Nos or bnl expression in these tissues. These results suggest that gut-derived Upd3 can control fat body/oenocyte expression of hypoxia regulators through both direct and indirect (relay) mechanism There is precedent for indirect, relay in the context of other Upd3/Upd2-mediated inter-organ responses. For example, in response to CO2, neuronal Upd3 controls blood cell differentiation in the lymph gland; however, this effect is not direct - Upd3 first signals to the fat body to induce Dilp6 expression, and Dilp6 then signals to the lymph gland to regulate hematopoiesis. A second example involves gut-derived Upd2: upon infection, Upd2 controls olfactory behavior, but does so via a relay in which Upd2 signals to glial cells, which in turn alter apolipoproteins expression, and these then modify olfactory neuron function.

    We have incorporated the new tissue-specific data into the manuscript and expanded the Discussion to address both direct and indirect modes of Upd3 action. (Fig 5 and lines 427-441)

    1. The authors relate the HIF-dependent limitation of Upd3 induction in hypoxia to regulation of cytokine-dependent immune responses in mammals; specifically they propose a parallel with a cytokine storm. This relationship is unclear to this reviewer, as in the Drosophila response Upd3 fulfils a signalling function (rather than immunological). I suggest they consider modifying this assumption.

    We appreciate this comment. Our intent in drawing a comparison to mammalian cytokine storm response was to illustrate the concept of fine-tuning cytokine responses, where too little or too much signaling can be deleterious, as we observe when comparing upd3 mutants to upd3-overexpressing animals. We have revised the Discussion to retain this concept while tempering the suggestion that our findings directly mirror cytokine storm pathologies in human (lines 511-536).

    1. Mitigation of lethality of HIF KD flies in Upd3 heterozygotes is very modest. Thus, the conclusion that one of the mechanisms by which HIF mediates adaptation to hypoxia is through inhibition of Upd3 expression is not sufficiently supported by the data. It seems like an over-interpretation of the results.

    We agree that the rescue is modest, and we would argue this may be expected given HIF-1's role as a master regulator that coordinates many gene expression changes required for hypoxia tolerance. Loss of HIF-1 therefore likely disrupts multiple essential processes simultaneously - including metabolic reprogramming and tracheal remodeling - that may not be restored by reducing upd3 dosage. We take the reviewer's point that this should not be framed as a primary mechanism. The partial reversal of lethality in upd3 heterozygotes nonetheless implicates excessive Upd3 signaling as one small component of what HIF-1 does to promote hypoxia adaptation, and we have revised the manuscript language to reflect this more measured interpretation (lines 529-536).

    1. HIF expression is well-known to reduce ROS levels in hypoxia by controlling mitochondrial activity through a wide array of mechanisms. Thus, this reviewer feels that the experiments utilized to rule out a role of ROS in fat body-to-gut communication are insufficient. Catalase reduces hydrogen peroxide levels, but not necessarily other reactive oxygen species. The authors might try to express other ROS scavengers such as superoxide dismutase. In addition, expression of scavengers should be carried out both at the fat body and gut.

    We thank the reviewer for this important point. We have now addressed it by overexpressing CatA, SOD1, or SOD2 individually in either fat body or enterocytes and measuring hypoxia-induced upd3 expression in each case. In all six conditions, hypoxia-induced upd3 expression was unaffected (Figs. S6B–G). Together, these experiments scavenge both hydrogen peroxide and superoxide in both tissues and collectively argue against a role for ROS in mediating upd3 induction

    __MINOR CONCERNS __

    1. The authors state that hypoxic upregulation of Upd3 in the gut occurs mostly in "large epithelial enterocytes". In Figure 3B, it is evident that GFP does not express in all cells; please utilize cell-type specific markers to identify which cells do express the cytokine.

    We appreciate this suggestion. Despite multiple requests to different laboratories, we were unable to obtain antibodies suitable for marking enterocyte subtypes in this context. To address the question of cell identity genetically, we used drivers specific for enterocytes (mex-GAL4) or progenitor cells (stem cells and enteroblasts; esg-GAL4) to drive RNAi-mediated knockdown of upd3 and then measured the effect on hypoxia-induced upd3 expression in whole guts. These experiments indicate that hypoxia-induced upd3 expression occurs mostly in enterocytes, with a smaller contribution from progenitor cells. This mirrors previous findings showing that infection-induced upd3 induction occurs in both enterocytes and enteroblasts, and supports our conclusion that enterocytes are the predominant source of hypoxia-induced Upd3. We have incorporated these results into the revised manuscript (Fig 3C and Fig S2C).

    1. The title of Fig 4 caption reads "Gut-derived upd3 controls adipose expression of hypoxia regulators." Only one hypoxia regulator has been analysed: Nitric Oxide Synthase. Please change the title to "Gut-derived upd3 controls adipose expression of Nitric Oxide Synthase."

    In the revised manuscript we now show that gut-derived Upd3 controls the expression of Nos, bnl, and Hipk in fat body and oenocytes, and that all three genes are required for hypoxia tolerance. We have therefore revised the figure title, to better reflect the findings presented in this version.

    1. Supplementary Figures 1 A and B lack statistical analysis.

    We have now included the statistical analyses in the revised manuscript figures.

    Reviewer 2

    __(Evidence, reproducibility and clarity (Required)): __This study by Ding and colleagues identifies a novel role for the cytokine Unpaired-3 (upd3) and the JAK/STAT signaling pathway coordinate a whole-body response to systemic hypoxia in Drosophila. The authors describe how low-oxygen conditions rapidly induce upd3 expression in both larvae and adults. Interestingly, this pathway's importance is sex-specific, as female flies require upd3 for survival in hypoxia, while males do not.

    Intriguingly, the authors identify the intestine as a crucial source of the hypoxia-induced upd3. This gut-derived upd3 then signals to the fat body and oenocytes, promoting the expression of nitric oxide synthase, which is essential for hypoxia tolerance. Furthermore, the study reveals an unexpected role for the transcription factor HIF-1α/sima as a molecular brake. Instead of simply promoting the hypoxia response, sima prevents the overproduction of upd3, demonstrating that a precise dosage of this cytokine is necessary for survival. The findings define a novel gut-to-fat/oenocyte signaling axis that coordinates systemic hypoxia adaptation and highlights the fly as an ideal system for studying interorgan communication during bouts of hypoxia. Overall, I find this manuscript an important step forward in understanding the link between hypoxia signaling and inflammation.

    __ (Significance (Required)): __This study is of high significance, as it not only demonstrates that a clear role for cytokine signaling in the Drosophila hypoxia response, but also demonstrates this response requires interorgan communication between adipose tissue and the intestine. Moreover, the study reveals a clear role for Hif1alpha in modulating upd3 expression, suggesting that this highly conserved transcription factor play a key role in fine tuning the inflammatory response.

    I think these findings are of broad interest and are potentially relevant to two aspects of public health. First, I believe the findings should be of particular interest to anyone studying hypoxic injuries, such as stroke and ischemia-reperfusion. Secondly, the observations could be relevant to a previous study that revealed an important role for hypoxia signaling in the mosquito larval intestine. Thus, this study could be important for revealing new mechanisms for inhibiting mosquito development, which would be of broad public health interest.

    Finally, I would highlight how this study raises a number of important question. Why are there sex-specific differences for upd3 in the hypoxia response? What is the signal from the fat body to the intestine? How does sima modulate upd3 signaling. Thus, I think this manuscript represents a foundation study that will be the motivation for numerous high-impact papers in the future.__ ____ __ We thank Reviewer 1 for their careful reading and comments on the manuscript. We are pleased that they found this to be "a study of high significance” that will be importance for our understanding of hypoxia and health. We have addressed each of their concerns in the revised manuscript as outlined below.

    __Major Concerns and Suggestions: __ I have no real for the manuscript as written - the experiments are well designed and control, the results, as presented, support the major conclusions. While there are clearly open questions, including what it the basis of the sex-specific effects, how does sima modulate upd3 expression, and what is the signal communicating fat body sima activity with intestinal upd3 expression, these open questions do NOT diminish the importance of the study.

    My only major concern is that the current draft lacks a discussion of previous studies in the mosquito Aedes aegypti, where hypoxia signaling plays a key role in larval development (https://doi.org/10.1073/pnas.1719063115). This body of literature should be incorporated into the discussion, as it hints at a conserved molecular mechanism.

    We thank the reviewer for pointing us to this important study. Valzania et al. demonstrate that gut hypoxia acts as a systemic signal in Aedes aegypti larvae, activating HIF to coordinate fat body metabolism and whole-body growth. We agree this is relevant context for our findings, as both studies support the idea that the gut can function as a hypoxia sensor that controls whole-body physiology through effects on the fat body. We have incorporated this into our Discussion (lines 488-492).

    Minor comments:

    Please include a list of fly stocks used in the methods with complete genotypes. Whenever possible, include the RRID number for the stock - these can be found on the BDSC page for the stock.

    We have now added the list of fly stocks as well as a supplemental table with full genotypes.

    Line 477-479 - provide citations that sima regulates glycolysis in the fly.

    We have now added these citations

    Lines 501-505 - please state if gasses were premixed or mixed in lab. Also, were flies contained in standard food vials during the exposure?

    We have now provided more detail on these points – the gases were premixed and flies were on standard food vials during the exposure.

    Lines 507-513 - how long after the hypoxia exposure were the flies assayed?

    We have now provided more detail on this point in the methods (lines 592-596) – the flies were assessed 24hrs after hypoxia exposure.

    In figures that display qRT-PCR data, please note that data were normalized to reference genes listed in Table S2.

    We have now added this methodological point.

    Please reference Flybase in either the acknowledgements or methods and include citations to the latest Flybase papers published in Genetics.

    We have now acknowledged Flybase and referenced the relevant papers

    Genetics nomenclature is inconsistent throughout the study, a few examples included: Figure legend 1 - italicize gene names Figure 2 legend - italicize upd3-null Line 259 - Capitalize gal4 Figure 4 legend - NOS is written in all capital, but in line 270, written as Nos. Please be consistent. Line 297 - gal4 is lower case, in contrast with elsewhere.

    We have now made these corrections

    Additional suggestions:

    While not required for publication, it would be interesting to examine intestinal upd3 expression when sima is inappropriately stabilized in the fat body of animals under normoxic conditions. This could be achieved by driving a fatiga-RNAi construct within the fat body.

    We did carry out this experiment but didn’t see any effect of fat body fatiga RNAi on gut upd3 levels.

    Reviewer 3

    __Evidence, reproducibility and clarity (Required)): Summary: While local cellular and organ adaptations to hypoxia are well-documented, organism-wide responses to systemic hypoxia are still not well understood. In this paper, the writers were interested in investigating how organisms adapt to systemic hypoxia. From their investigations, they were able to show that gut-derived upd3 is crucial to animals' tolerance to hypoxia. They also show that the master hypoxia regulator Sima is required to keep the upd3 level in check to avoid the deleterious effect of excess upd3. They also showed that the fatbody Sima is important in the regulation of gut-upd3 level, showing an inter-organ communication network in the adaptation to systemic hypoxia. One of their findings shows sex dimorphism in hypoxia tolerance; however, they did not show the mechanism behind this. I think the major weakness is not knowing how the animal actually fail to survive. What causes reduced survival should be explored. Generally, the studies show how animals adapt to systemic hypoxia, this knowledge is important in systemic hypoxia pathology.

    __

    __Significance (Required)): __This paper explores how the organism copes with hypoxia, and explored how Upd from the gut plays a role in mediating this response in the fat body and the oenocytes

    We thank Reviewer 1 for their careful reading and comments on the manuscript. We have addressed each of their concerns in the revised manuscript as outlined below.

    __Major comment: __

    Figure 1: The authors clearly showed that Upd3 level was up in the hypoxia condition and is important for animal tolerance to hypoxia. Apart from Upd3, are there other members of the unpaired family increasing and involved in hypoxia tolerance?

    We thank the reviewer for this question. We examined expression of all three unpaired family members and found that both upd2 and upd3 are induced by hypoxia, while upd1 is not. We also have preliminary evidence that upd2 mutants show reduced hypoxia survival, and that this effect is not additive with loss of upd3. While these early results are intriguing, this paper is focused on defining the role of upd3 in hypoxia tolerance, and exploring upd2, both alone and in combination with upd3, across different aspects of hypoxia biology we see as the basis of future investigations.

    Notably, co-induction of upd2 and upd3 by the same stress is a recurring theme in Drosophila biology, yet their respective contributions to organismal physiology are complex - sometimes overlapping, sometimes distinct - and in many studies only one family member has been characterized in detail. Indeed, our current understanding of how upd2 and upd3 each contribute to responses to infection, high-fat diet, and other stresses has emerged from the collective findings of multiple independent studies rather than from any single paper addressing both cytokines simultaneously. For example, during infection both Upd2 and Upd3 are induced in the gut to promote stem cell-mediated repair, yet only Upd2 has been shown to additionally signal to the brain to control olfactory behavior. Similarly, on a high-fat diet both cytokines are upregulated, but with distinct effects on different aspects of organismal biology: enterocyte-derived Upd3 promotes intestinal stem cell divisions, hemocyte-derived Upd3 controls fat body lipid levels, and fat body-derived Upd2 alters nephrocyte function. We see the current study as a foundation for broader investigations into unpaired cytokine biology in hypoxia. Indeed, Reviewer 2 noted that this manuscript "represents a foundation study that will be the motivation for numerous high-impact papers in the future", and we anticipate that the effects of Upd2 and Upd3 in hypoxia will prove similarly pleiotropic and resolving their respective contributions to different aspects of organismal biology in low oxygen will require dedicated future investigation.

    Figure 2: From the method, female and male flies were subjected to different durations of hypoxia, 24-28 hours for females and 16-18 hours for males. What happens when subjecting different sexes to similar periods of hypoxia?

    We thank the reviewer for this question. Males and females show inherently different sensitivities to hypoxia, as they do for other environmental stresses such as starvation. To reliably detect genetic effects on hypoxia tolerance, it is important to use exposure conditions that produce partial lethality in controls (50-80% survival), ensuring experiments are conducted within the appropriate range of hypoxic sensitivity for each sex. Because males and females differ in their sensitivity, no single timepoint satisfies this criterion for both sexes. When males are exposed for the same duration used in female experiments (24-28h), all animals - controls and experimental genotypes alike - die, precluding any meaningful comparison. Conversely, exposing females to the shorter timepoint used for males (16-18h) produces no detectable lethality, making it equally uninformative. The sex-specific exposure durations we use are therefore an experimental design choice that allows us to assess hypoxia tolerance appropriately in each sex.

    Upon concluding that gut derived upd affects fat and oenocytes, it is a bit strange that the qPCR is done in the abdomen, which is presumably where the gut is. Should the gut be excluded in these assays?

    We thank the reviewer for raising this point. For abdominal qRT-PCR experiments examining fat body and oenocyte gene expression, we dissected and removed the gut and ovaries prior to RNA extraction, leaving an abdominal sample enriched in fat body and oenocytes. We have clarified this in the Methods and Results section of the revised manuscript (Lines 245-246 and 626-627).

    It is important to establish how the animals die under hypoxia.

    We thank the reviewer for raising this important question. Our results show that gut-derived Upd3 is required for hypoxia tolerance in part through its control of Nos, bnl, and Hipk expression in fat body and oenocytes, and that knockdown of each of these genes individually reduces hypoxia survival. However, precisely why animals die when upd3 or these downstream effectors are lost remains an open question, and we discuss much of what we outline below in the revised manuscript Discussion (lines 443-466).

    All three effectors are signaling molecules, and we speculate that they likely coordinate further downstream processes required for hypoxia tolerance, either within fat body and oenocytes or by acting on other tissues. In particular, both bnl, an FGF ligand, and nitric oxide, produced downstream of Nos, have established roles in tracheal development and remodeling, raising the possibility that Upd3-dependent regulation of tracheal responses to hypoxia contributes to survival. Nitric oxide can also regulate nitrosylation and has been shown to affect the unfolded protein response, a conserved pathway induced by hypoxia. bnl, in addition to its role in tracheal remodeling, has been shown to regulate metabolic changes in target tissues. Hipk is a kinase with likely many downstream targets and has been shown in flies to control metabolism and mitochondrial function. Together, these observations suggest that Upd3 engages a broad downstream signaling network, the full scope of which remains to be defined.

    We think this situation is analogous to other environmental stresses such as starvation, where survival requires the coordinated regulation of a spectrum of physiological processes across multiple tissues, and where even well-characterized regulators are known to engage many downstream targets and pathways. We see the current paper as establishing the gut-to-fat body Upd3 requirement for hypoxia tolerance, and we suggest this lays a foundation for future exploration of the full spectrum of Upd3 targets and investigation of how they coordinate adaptive responses to low oxygen.

    Figure 3-6: Controls for RNAi experiments - is there any reason for not using RNAi-specific control, such as mcherry-RNAi, lacZ-RNAi, etc, rather than a wildtype control in all the RNAi-mediated knockdowns? Please address this. Don't necessarily have to repeat all the experiments using RNAi-specific control, but repeating just a few to show that both wild-type and UAS-RNAi-specific controls show similar results would be important.

    We thank the reviewer for raising this point. To address potential non-specific effects of RNAi expression on hypoxia tolerance, we expressed control GFP RNAi or mCherry RNAi transgenes using the main Gal4 drivers employed in this study: mex-Gal4 (gut) and desat;r4-Gal4 (fat body and oenocytes), and found no effect on hypoxia survival compared to wild-type controls (Fig S2E and S4B). These results indicate that RNAi expression per se does not adversely affect hypoxia tolerance, and that the survival effects we observe reflect specific knockdown of the genes of interest.

    Although gut-derived upd3 contributes largely (40%) to hypoxia tolerance, what other tissues' upd3 is important for hypoxia tolerance?

    We thank the reviewer for this important question. We find that upd3 is induced in multiple tissues during hypoxia, including the head, thorax, and abdomen. However, when we knocked down upd3 using drivers targeting the major cell types in these tissues, including muscle, neurons, and fat body/oenocytes, we observed no significant effect on hypoxia survival, in contrast to the robust effect seen with gut-specific knockdown. These new data, included in the revised manuscript, suggest that gut-derived Upd3 is a primary contributor to hypoxia tolerance (Fig S3).

    That said, we do not conclude that the gut is the only relevant source. Other tissues we have not yet examined, including hemocytes, glia, and tracheal cells, may also contribute, and it is possible that Upd3 produced from multiple tissues acts redundantly, such that knockdown in any single tissue other than the gut is insufficient to cause a survival defect. By analogy with other stress contexts such as nutrient deprivation and infection, where upd cytokines are produced from multiple tissues and exert distinct effects on different aspects of physiology, we anticipate that Upd3 from tissues other than the gut may well contribute to hypoxia tolerance. However, fully defining these contributions will require detailed tissue-specific experiments that are beyond the scope of the current paper and will be the focus of future investigations. We have expanded on this point in the Discussion of the revised manuscript (lines 420-425).

    Can you use a hypoxia readout to experimentally show that the gut is the main sensor of hypoxia compared to other tissues? Looking at the data, the fatbody could also be major sensors of hypoxia. Therefore, investigating hypoxia readout in these and other tissues would further strengthen the direction of communication.

    We thank the reviewer for this suggestion, however, we wish to clarify that we are not claiming the gut is the main or primary sensor of hypoxia. All tissues are likely capable of sensing low oxygen and mounting cell-autonomous responses, and in some cases perhaps also non-autonomous signals to other tissues. Our findings specifically show that one consequence of gut hypoxia sensing is upregulation of Upd3, which then acts as an inter-organ signal to coordinate responses in target tissues such as the fat body and oenocytes. The fat body itself also senses hypoxia and mounts its own responses, as we and others have shown, including HIF-dependent regulation of gut Upd3 expression described in this paper. An analogous situation exists during nutrient starvation, where all cells autonomously sense and respond to nutrient deprivation, but on top of these cell-autonomous responses, specific tissues also mediate inter-organ signaling to coordinate whole-body physiological adaptations. We propose that hypoxia responses are organized similarly, and that the gut-to-fat body Upd3 signaling axis we describe here represents one such inter-organ communication pathway. We have clarified this point in the revised manuscript (lines 468-492).

    __Minor comment:

    __

    Should check the alignment of the confocal image in Figure 3b, especially the top panel.

    We have now fixed the images to better align them

    Figure 6: "gut-specific sima knockdown (mex>sima-RNAi) did not significantly alter intestinal upd3 mRNA levels compared to controls (mex>+) under hypoxic conditions (Figure 6C)." This statement refers to Figure 6B, not Figure 6C

    We have now corrected this

    Since the fat body Sima non-autonomously control the gut upd3 level, can you also show this functionally important by investigating the animal's survival or other functional studies?

    We thank the reviewer for this suggestion. Ideally, we would manipulate sima and upd3 independently and in parallel, knocking down sima specifically in the fat body while simultaneously reducing upd3 in the gut, to directly test the functional importance of this inter-organ axis for survival. In principle this could be achieved using orthogonal binary expression systems such as the GAL4/UAS and QF/QUAS systems in combination, but this would require the development of new genetic tools. An additional challenge is that based on our results, such experiments would require fine-tuned reduction of gut upd3, sufficient to suppress the elevated levels caused by fat body sima knockdown, but not so low as to itself compromise survival, as we have shown that loss of upd3 is detrimental. For these reasons, while we agree these would be, in principle, interesting experiments, they would technically be challenging to carry out.

    Strangely, all the statistically significant data/results from both supplementary and main figures had a one-star significance even in graphs with very obvious differences and less sample variation.

    We thank the reviewer for this observation. In all figures, a single asterisk is used to denote statistical significance at p < 0.05, regardless of whether the actual p value is substantially lower. This is a presentation convention we adopted consistently across all figures rather than a reflection of the strength of the underlying differences.

  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

    Summary: While local cellular and organ adaptations to hypoxia are well-documented, organism-wide responses to systemic hypoxia are still not well understood. In this paper, the writers were interested in investigating how organisms adapt to systemic hypoxia. From their investigations, they were able to show that gut-derived upd3 is crucial to animals' tolerance to hypoxia. They also show that the master hypoxia regulator Sima is required to keep the upd3 level in check to avoid the deleterious effect of excess upd3. They also showed that the fatbody Sima is important in the regulation of gut-upd3 level, showing an inter-organ communication network in the adaptation to systemic hypoxia. One of their findings shows sex dimorphism in hypoxia tolerance; however, they did not show the mechanism behind this. I think the major weakness is not knowing how the animal actually fail to survive. What causes reduced survival should be explored. Generally, the studies show how animals adapt to systemic hypoxia, this knowledge is important in systemic hypoxia pathology.

    Major comment:

    • Figure 1: The authors clearly showed that Upd3 level was up in the hypoxia condition and is important for animal tolerance to hypoxia. Apart from Upd3, are there other members of the unpaired family increasing and involved in hypoxia tolerance?
    • Figure 2: From the method, female and male flies were subjected to different durations of hypoxia, 24-28 hours for females and 16-18 hours for males. What happens when subjecting different sexes to similar periods of hypoxia?
    • Upon concluding that gut derived upd affects fat and oenocytes, it is a bit strange that the qPCR is done in the abdomen, which is presumably where the gut is. Should the gut be excluded in these assays?
    • It is important to establish how the animals die under hypoxia.
    • Figure 3-6: Controls for RNAi experiments - is there any reason for not using RNAi-specific control, such as mcherry-RNAi, lacZ-RNAi, etc, rather than a wildtype control in all the RNAi-mediated knockdowns? Please address this. Don't necessarily have to repeat all the experiments using RNAi-specific control, but repeating just a few to show that both wild-type and UAS-RNAi-specific controls show similar results would be important.
    • Although gut-derived upd3 contributes largely (40%) to hypoxia tolerance, what other tissues' upd3 is important for hypoxia tolerance?
    • Can you use a hypoxia readout to experimentally show that the gut is the main sensor of hypoxia compared to other tissues? Looking at the data, the fatbody could also be major sensors of hypoxia. Therefore, investigating hypoxia readout in these and other tissues would further strengthen the direction of communication.

    Minor comment:

    • Should check the alignment of the confocal image in Figure 3b, especially the top panel.
    • Figure 6: "gut-specific sima knockdown (mex>sima-RNAi) did not significantly alter intestinal upd3 mRNA levels compared to controls (mex>+) under hypoxic conditions (Figure 6C)." This statement refers to Figure 6B, not Figure 6C
    • Since the fat body Sima non-autonomously control the gut upd3 level, can you also show this functionally important by investigating the animal's survival or other functional studies?
    • Strangely, all the statistically significant data/results from both supplementary and main figures had a one-star significance even in graphs with very obvious differences and less sample variation.

    Significance

    This paper explores how the organism copes with hypoxia, and explored how Upd from the gut plays a role in mediating this response in the fat body and the oenocytes

  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

    This study by Ding and colleagues identifies a novel role for the cytokine Unpaired-3 (upd3) and the JAK/STAT signaling pathway coordinate a whole-body response to systemic hypoxia in Drosophila. The authors describe how low-oxygen conditions rapidly induce upd3 expression in both larvae and adults. Interestingly, this pathway's importance is sex-specific, as female flies require upd3 for survival in hypoxia, while males do not.

    Intriguingly, the authors identify the intestine as a crucial source of the hypoxia-induced upd3. This gut-derived upd3 then signals to the fat body and oenocytes, promoting the expression of nitric oxide synthase, which is essential for hypoxia tolerance. Furthermore, the study reveals an unexpected role for the transcription factor HIF-1α/sima as a molecular brake. Instead of simply promoting the hypoxia response, sima prevents the overproduction of upd3, demonstrating that a precise dosage of this cytokine is necessary for survival. The findings define a novel gut-to-fat/oenocyte signaling axis that coordinates systemic hypoxia adaptation and highlights the fly as an ideal system for studying interorgan communication during bouts of hypoxia. Overall, I find this manuscript an important step forward in understanding the link between hypoxia signaling and inflammation.

    Major Concerns and Suggestions:

    I have no real for the manuscript as written - the experiments are well designed and control, the results, as presented, support the major conclusions. While there are clearly open questions, including what it the basis of the sex-specific effects, how does sima modulate upd3 expression, and what is the signal communicating fat body sima activity with intestinal upd3 expression, these open questions do NOT diminish the importance of the study.

    My only major concern is that the current draft lacks a discussion of previous studies in the mosquito Aedes aegypti, where hypoxia signaling plays a key role in larval development (https://doi.org/10.1073/pnas.1719063115). This body of literature should be incorporated into the discussion, as it hints at a conserved molecular mechanism.

    Minor comments:

    Please include a list of fly stocks used in the methods with complete genotypes. Whenever possible, include the RRID number for the stock - these can be found on the BDSC page for the stock.

    Line 477-479 - provide citations that sima regulates glycolysis in the fly.

    Lines 501-505 - please state if gasses were premixed or mixed in lab. Also, were flies contained in standard food vials during the exposure?

    Lines 507-513 - how long after the hypoxia exposure were the flies assayed?

    In figures that display qRT-PCR data, please note that data were normalized to reference genes listed in Table S2.

    Please reference Flybase in either the acknowledgements or methods and include citations to the latest Flybase papers published in Genetics.

    Genetics nomenclature is inconsistent throughout the study, a few examples included:

    Figure legend 1 - italicize gene names

    Figure 2 legend - italicize upd3-null

    Line 259 - Capitalize gal4

    Figure 4 legend - NOS is written in all capital, but in line 270, written as Nos. Please be consistent.

    Line 297 - gal4 is lower case, in contrast with elsewhere.

    Additional suggestions:

    While not required for publication, it would be interesting to examine intestinal upd3 expression when sima is inappropriately stabilized in the fat body of animals under normoxic conditions. This could be achieved by driving a fatiga-RNAi construct within the fat body.

    Significance

    This study is of high significance, as it not only demonstrates that a clear role for cytokine signaling in the Drosophila hypoxia response, but also demonstrates this response requires interorgan communication between adipose tissue and the intestine. Moreover, the study reveals a clear role for Hif1alpha in modulating upd3 expression, suggesting that this highly conserved transcription factor play a key role in fine tuning the inflammatory response.

    I think these findings are of broad interest and are potentially relevant to two aspects of public health. First, I believe the findings should be of particular interest to anyone studying hypoxic injuries, such as stroke and ischemia-reperfusion. Secondly, the observations could be relevant to a previous study that revealed an important role for hypoxia signaling in the mosquito larval intestine. Thus, this study could be important for revealing new mechanisms for inhibiting mosquito development, which would be of broad public health interest.

    Finally, I would highlight how this study raises a number of important question. Why are there sex-specific differences for upd3 in the hypoxia response? What is the signal from the fat body to the intestine? How does sima modulate upd3 signaling. Thus, I think this manuscript represents a foundation study that will be the motivation for numerous high-impact papers in the future.

  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

    The authors describe a role of Unpaired 3 (Upd3) in tissue communication in responses to hypoxia in Drosophila adult flies. Upd3 mRNA is strongly upregulated in hypoxia, along with well-characterized JAK/STAT downstream target genes, in both adult fly males and females, as well as in larvae. Interestingly, adult females but not males require Upd3 for 15 to 24 h survival in hypoxia, as Upd3 mutant females but not males die to a much larger proportion in these conditions. Adult females they display strong hypoxic upregulation of Upd3 in the gut, assessed by RT-PCR or through a Gal4 transcriptional reporter, mainly in epithelial enterocytes. Enterocyte-specific RNAi-mediated KD indicated that this enterocyte expression of Upd3 represents about 40% of Upd3 expression in the whole body. Enterocyte-specific KD of Upd3 in adult females significantly reduced survival in hypoxia, suggesting that this expression is critical for hypoxic adaptation. Tissue-specific analysis of the expression of the STAT target genes, SOCS36E, TotA and TotM revealed that stimulation of the JAK/STAT pathway in hypoxia is widespread, although more pronounced in abdominal tissues. Indeed, overexpression of Upd3 in enterocytes provokes upregulation o both target genes TotA and TotM. Consistent with this RNAi-dependent inhibition of the JAK/STAT pathway in the fat body and oenocytes significantly reduced survival of female flies in hypoxia. Nitric oxide synthase (NOS) is strongly upregulated in adult female abdomens upon hypoxic exposure, and KD of NOS in fat body and oenocytes reduced hypoxic survival. Surprisingly, the found that ubiquitous KD of HIFa/Sima led to mitigation of Upd3 hypoxic induction and, more clearly, to JAK/STAT target gene induction. HIF KD flies displayed increased lethality in hypoxia, and this lethality was slightly mitigated in Upd3 heterozygous flies. The authors conclude that increased lethality of HIF-minus flies in hypoxia stems at least in part from excessive levels of Upd3. The authors then find that HIF/Sima-dependent inhibition of Upd3 expression is non-cell autonomous, since KD of Sima specifically in the gut does not affect expression of Upd3 in this organ. Instead, Sima KD at the fat body led to significant increase of Upd expression in the gut, suggesting that a Sima-born signal communicates these two organs, leading to restriction of Upd3 intestinal expression. ROS does not seem to be the signal that communicates the fat body with the gut, as expression of catalase in the fat body did not affect expression of Upd3 in the gut.

    Major concerns

    1. Figure 1 lacks statistical analysis. It is important to determine if the apparent differences in gene expression are statistically significant.

    2. Is NOS expression in fat body/oenocytes JAK/STAT-dependent? Block the pathway in hypoxia specifically in this cells and check.

    3. The authors relate the HIF-dependent limitation of Upd3 induction in hypoxia to regulation of cytokine-dependent immune responses in mammals; specifically they propose a parallel with a cytokine storm. This relationship is unclear to this reviewer, as in the Drosophila response Upd3 fulfils a signalling function (rather than immunological). I suggest they consider modifying this assumption.

    4. Mitigation of lethality of HIF KD flies in Upd3 heterozygotes is very modest. Thus, the conclusion that one of the mechanisms by which HIF mediates adaptation to hypoxia is through inhibition of Upd3 expression is not sufficiently supported by the data. It seems like an over-interpretation of the results.

    5. HIF expression is well-known to reduce ROS levels in hypoxia by controlling mitochondrial activity through a wide array of mechanisms. Thus, this reviewer feels that the experiments utilized to rule out a role of ROS in fat body-to-gut communication are insufficient. Catalase reduces hydrogen peroxide levels, but not necessarily other reactive oxygen species. The authors might try to express other ROS scavengers such as superoxide dismutase. In addition, expression of scavengers should be carried out both at the fat body and gut.

    Minor concerns

    1. The authors state that hypoxic upregulation of Upd3 in the gut occurs mostly in "large epithelial enterocytes". In Figure 3B, it is evident that GFP does not express in all cells; please utilize cell-type specific markers to identify which cells do express the cytokine.

    2. The title of Fig 4 caption reads "Gut-derived upd3 controls adipose expression of hypoxia regulators." Only one hypoxia regulator has been analysed: Nitric Oxide Synthase. Please change the title to "Gut-derived upd3 controls adipose expression of Nitric Oxide Synthase."

    3. Supplementary Figures 1 A and B lack statistical analysis.

    Significance

    This is an important study, because most previous studies have focused on cell-autonomous responses to hypoxia, but much less is known about systemic responses to low oxygen conditions, particularly in relation to inter-organ communication during this responses. This work defines the cytokine unpaired 3, homolog of human interleukin 6, as a major regulator of systemic responses to hypoxia. Future studies will determine if interleukin 6 plays similar roles in mammals. This work might be of interest for a broad audience interested in responses to hypoxia, as well as general physiology.

  5. PREreview

    This Zenodo record is a permanently preserved version of a PREreview. You can view the complete PREreview at https://prereview.org/reviews/17804333.

    PREreview of "Cytokine control of systemic hypoxia tolerance in Drosophila".

    By Rebecca S. Moore of HHMI Transparent and Accountable Peer Review Training Pilot.

    Ding et al. reveals that Unpaired-3 (upd3)- HIF-1a/sima mediates systemic hypoxia tolerance in Drosophila through a sex-specific gut-to-fat body/oenocyte signaling axis that regulates nitric oxide synthase expression.

    This paper addresses several knowledge gaps leading to advancements in the field. Although gut-to-fat/oenocyte signaling has been documented in Drosophila, the context in which it functions was previously unknown. Additionally, HIF-1a/sima activation has been well studied as an adaptive response to hypoxia and stress, however the role of sima as a negative regulator of its own transcriptional activation has not been known. Lastly, while sex-specific differences in stress responses have been documented, specific requirements for cytokine signaling pathways in one sex for survival under hypoxia was unknown. This finding points to fundamental differences in how male and females physiologically adapt to low-oxygen stress.

    Specifically, Ding et al identifies a previously unknown interorgan communication pathway where the cytokine upd3 is produced in the gut in response to hypoxia. Gut derived-upd3 then signals to the fat body/oenocytes to promote survival. The authors provide evidence that HIF-1a/sima functions in the fat body, acting as a non-cell autonomous molecular brake preventing lethal upd3 overproduction in the gut. This finding presents a more nuanced role for sima, suggesting it not only activates hypoxia response genes but also prevents detrimental hyper-inflammatory responses. Finally, the authors demonstrate a sex-specific requirement for upd3 in hypoxia.

    The authors conclusions are supported, but there are areas in which the paper can be made clearer and/or additional experiments can further strengthen their findings.

    General Major Points

    1. The authors use one RNAi or mutant line in all experiments. Using additional RNAi lines or mutant alleles would help to support their data. This would help eliminate background effects and show reproducibility across genetic contexts.

    2. The authors claims would benefit from rescue experiments (i.e. rescue of upd3 or STAT92E in upd3 mutants) to show sufficiency and complete the causal chain of events – hypoxia -> high upd3 in gut -> sima activity in fat body/oenocytes -| hyper activation of upd3 in gut.

    Specific Major Points

    The claim that hypoxia rapidly induces upd3 and JAK/STAT signaling is supported by consistent time-course qRT-PCR data showing increased upd3 and multiple STAT targets across sexes and developmental stages (Figure 1), and by two enterocyte drivers (Figure 3C-F). However, reliance on mRNA readouts without direct upd3 protein/secreted cytokine measures or canonical STAT activation assays leaves a gap between transcriptional induction and bona fide pathway activation. 

    1. Use UAS STAT92E RNAi to test whether hypoxia-induced SOCS36E and Turandot transcripts require STAT92E in qRT-PCR experiments. The authors should perform qRT-PCR experiments for SOCS36E/Turandot mRNA levels under hypoxia conditions with STAT92E knocked down in the fat body or oenocytes. This will provide pathway-dependency evidence that transcript induction reflects JAK/STAT activation.

    2. Add direct pathway activation readouts and cytokine measurements using GFP/LacZ-type reporters and quantify hypoxia-induced reporter activity in gut and abdominal tissues via microscopy or use immunofluorescence of phosphorylated STAT to assess nuclear accumulation under hypoxia. These experiments can be validated in a upd3 mutant. These experiments would validate JAK/STAT pathway activation rather than relying on mRNA readouts.

    The evidence for abdominal JAK/STAT activation under hypoxia and induction of STAT target genes in the abdomen upon enterocyte upd3 overexpression is compelling at the whole tissue level (Figure 4B). However, these measurements were made on bulk abdomens and do not localize the response to oenocytes or the fat body.

    1. Perform tissue specific STAT necessity by repeating these readouts in r4>STAT92E RNAi or desat>STAT92E RNAi backgrounds to test whether enterocyte upd3 driven inductions require STAT in each tissue/cell type. Additionally, the authors can use an imaging-based approach to visualize upd3 activation in each tissue/cell type. This would provide direct evidence of cell-type specific activation and necessity.

    HIF-1a/sima functions as a molecular brake in the fat body that limits upd3 signaling in the gut during hypoxia. The authors provide 3 lines of evidence that support this claim including whole-body qRT-PCR and survival experiments (Figure 5 and 6). However, the key weakness includes the lack of direct validation of sima RNAi knockdown, absence of pathway epistasis or direct pathway activity showing that the sima-dependent activation is causally mediated by upd3-STAT signaling in the fat body/oenocytes. This would provide clear validation of the model.

    1. Perform genetic epistasis experiments to test whether gut upd3 knockdown in the gut suppresses JAK/STAT pathway activation in the fat body/oenocytes and improves hypoxia survival. These experiments would provide pathway-specific evidence that fat body sima-dependent cytokine amplification requires gut upd3 and JAK/STAT in fat body/oenocytes.

    2. Use reporter lines to image JAK/STAT activity in the tissues/cells of interest. These experiments would establish that sima restrains upd3 driven JAK/STAT signaling at the protein level.

    Minor Points

    1. In Figure 1A: the authors do not provide any statistical analysis of qRT-PCR data. Providing statistics would help readers understand if upd activation reaches a peak within the time frame studied or if it will continue to increase. Most commonly, qPCR statistics are run on DCt values and not relative expression as relative expression quantification is non linear. The authors can use a One-Way ANOVA followed by a Tukey's multiple comparison test to achieve statistical power.

    2. In Figure 1E-F: The finding that this mechanism is true in larvae and adult flies is interesting. There is missing information about the stage of the larvae that this was tested in. Providing clearer methods would help any scientist who wants to repeat and work on these findings.

    3. Figure 6B is called out as Figure 6C. Please update the correct figure call out to account for Figure 6B.

    4. The authors provide no information about an internal control in their qRT-PCR experiments. Internal controls are essential for normalization and would strengthen the findings that the expression changes identified are truly hypoxia responsive.

    5. Readers would benefit from visualizing the entire survival curves and use of Kaplan Mier statistics which are typical for survival assays. This would allow readers to see if the LD50 were different depending on environmental treatment or if one fly survived much longer to skew the survival statistics.

    6. It would be extremely helpful to have clarity in the methods about the survival assays done. More information would give readers clarity to repeat or interrupt data.

    Competing interests

    The author declares that they have no competing interests.

    Use of Artificial Intelligence (AI)

    The author declares that they did not use generative AI to come up with new ideas for their review.

  6. Version published to 10.1101/2025.06.30.661614 on bioRxiv