Ir para a Avaliação PREreview

Avalilação PREreview de Insulin-like peptide secretion is mediated by peroxisome-Golgi interplay

de Ink Pelican do/a HHMI Transparent and Accountable Peer Review Training Pilot
Publicado
DOI
10.5281/zenodo.17078493
Licença
CC BY 4.0

König et al. explore the relationship between peroxisomes and neuropeptides using the insulin-producing cells (IPCs) in Drosophila melanogaster. By utilizing a very clean assay it is evident that Dilp2 accumulates in some of the IPCs in conditions of starvation and is no longer after feeding. This release of neuropeptide is not present if the animal has loss-of- function mutations in genes key for peroxisome biogenesis or a loss-of-function mutation in proteins required for Golgi-peroxisome lipid transfer but is still present in animals with loss-of-function mutations in a peroxisome enzyme maintenance protein. This indicates that peroxisome biogenesis, as well as lipid transfer from the Golgi to peroxisomes, is required for neuropeptide secretion. The authors further support this using lipidomics. Interestingly, the authors also determine that neuronal activity is not altered by the presence of peroxisome biogenesis proteins, indicating a dissociation between neuronal activity and neuropeptide

The strengths of this paper are that their primary assay for monitoring Dilp2 release has a very strong and clean phenotype, and there is no doubt that they are monitoring the release of Dilp2. The absence then, of Dilp2 release in the pex19/pex3/wat-deficient animals is therefore very convincing that the aberrations to their normal function are impacting neuropeptide release. The authors build a compelling picture of the wide- ranging effects that impacting peroxisome biogenesis causes, such as the organism-wide altered lipid profile, the prevention of lipid release from the fat body and gut, and the increased retention of peroxisomal matrix proteins.

While the authors provide some compelling images showing the interactions of Golgi and peroxisome markers in fed, starved, and refed conditions, seeing the presence of Dilp2 in these peroxisomes would better associate this interaction as the reason neuropeptides are being retained during starvation and increase the strength of the conclusions of the paper. Furthermore, seeing these interactions impacted in the pex19/pex3 peroxisomal mutants vs those in the non-peroxisome biogenesis mutants pex2 would provide strong evidence that these interactions are causing the decrease in Dilp2 release post-feeding.

Points for the authors

Major Points

1.        The authors show that the peroxisomes and Golgi are near each other in Figure 5 and Supp Figure 3, and they argue in the discussion (line 405) that this interaction enables the secretion of insulin-like peptides. While they do show this interaction and show that neuropeptides are not released, more evidence is needed to support the claim that this interaction is the reason for this. By doing the experiments shown in Figure 3 with their antibody against Dilp2, they could support their claim by showing Dilp2 accumulation in peroxisomes at the Golgi. Additionally, showing that the Golgi-peroxisome connection across fed-states does not mean that this connection is what is aberrant in the peroxisome biogenesis mutant lines. Using the same experimental parameters used in the Figure 3 experiments, but in a pex19/pex3 mutant background to show the increase in Golgi-peroxisome connections, and in a pex2 mutant background to show no change in Golgi- peroxisome connections, would make their claim on line 404 very strong.

2.        The manuscript utilizes a variety of RNAi lines and antibodies against different proteins. However, it is unclear whether these reagents were validated, either in the current study or through referenced literature. Providing information on reagent efficiency and potential off-target effects would lend greater confidence to the study’s conclusions.

3.        A particularly exciting result is that neuronal activity appears uncoupled from neuropeptide release. Figure 2 shows that pex19 mutant animals exhibit “normal” activity levels following starvation and refeeding but do not release neuropeptides. However, this interpretation would benefit from comparison to control animals under the same conditions. Furthermore, in the standard diet condition for pex19 mutants, the activity distribution is difficult to interpret due to the clustering of data points near zero. Additional replicates or clearer quantification may help clarify the interpretation.

Minor Points

1.        In Figure 1A, it should be explicitly stated in the figure that the image depicts Dilp2 antibody staining so that the readers understand this is not a transgene.

2.        In Figure 1E, it would be helpful to elaborate on what occurs to pex19/3/16 during peroxisome maturation, and to clarify where Pex2/1 come from. Adding this to the schematic would improve clarity for the reader.

3.        The manuscript shows Dilp2 release from IPC somas, but it remains unclear whether the peptides are being directly secreted from the soma or trafficked along axons to synapses. Clarification on the route of peptide transport would strengthen the narrative.

4.        Line 143: Figure 2C is cited, but the corresponding data appears to be in Figure 1C. This should be corrected.

5.        Line 147: The hemocyte rescue of pex19 mutants appears more robust than the IPC rescue. Could the authors elaborate on this finding, especially since the role of pex19 has primarily been discussed in the context of IPCs?

6.        Lines 153–157: The term “depletion of peroxisomes” would benefit from clarification—does it refer to the loss of lipids, membrane proteins, matrix proteins, or all the above? The definition of matrix proteins (presumably lumenal proteins) should also be made more explicit. Additionally, the term “ghosts” is unclear—do these refer to empty peroxisomes, or to aggregates of membrane proteins? Further explanation would be appreciated to clarify what change is actually occurring within the peroxisomes.

7.        The same figure panel appears in both Figure 3A'' and Supplemental Figure 2A (chart 3, PI), which may lead to confusion. It should be shown in only one location to avoid redundancy.

8.        Line 263: BODIPY-Cer is referred to as a Golgi marker, yet later described as labeling the plasma membrane. This inconsistency should be addressed, and the appropriate designation clarified. Additionally, Figure 4E: The saturated images help highlight signal accumulation, but make it difficult to determine whether accumulation occurs at the Golgi. Including an unsaturated version of the image in the supplement would help address this.

9.        Line 292: The px-mCherry tool, as well as other transgenic Golgi markers, are novel and useful. However, it is important to demonstrate that they do not perturb Golgi- peroxisome interactions. Using the Dilp2 release assay (from Figure 1) in these transgenic backgrounds would provide reassurance.

10.  Figures 5A/C: It is not immediately clear which label corresponds to the Golgi and which to peroxisomes. A consistent color scheme across panels or individual legends for each would aid comprehension.

11.  In Figure 5, the Golgi appears to enlarge upon refeeding. If quantified, this result should be shown. If not, the authors may wish to consider quantification, as an increase in Golgi size could contribute to increased interaction probability.

12.  Line 363: The BioGRID database may not be familiar to all readers. A brief explanation of what BioGRID is and how it was used in this context would be helpful.

13.  Line 440: The manuscript states that GM130 accumulation in pex19 mutants suggests missorting of membrane proteins. However, as GM130 is a matrix protein, this conclusion seems inconsistent. It may be more accurate to suggest the mislocalization of matrix proteins or to revise this statement.

14.  Line 659: The 500nm threshold used to define dense-core vesicles (DCVs) appears quite large, as typical DCV sizes range from 100–150nm. The authors should clarify the rationale behind this threshold.

15.  The relevance of Figures 1E and 6G to Dilp2 biology is unclear, as the peptide is not shown. In Figure 1E, the diagram seems to depict peroxisome stages rather than function. Please revise to better illustrate Dilp2 biology or clarify their intended message.

Reviewer Disclosure:

Language in the Minor and Major comments of this review was refined with the assistance of OpenAI's ChatGPT to improve clarity and tone.

Competing interests

The author declares that they have no competing interests.

Comentários

Escrever um comentário
  1. de Nicole Kucharowski
    Publicado
    Licença
    CC BY 4.0

    Major Points 

    1. We appreciate the reviewer’s suggestion to apply Dilp2 antibody staining and to analyze Golgi-peroxisome contacts across multiple peroxisome biogenesis mutant backgrounds. However, two main points need clarification. First, our manuscript does not claim that Dilp2 accumulates within peroxisomes at the Golgi; rather, we propose that Golgi-peroxisome proximity facilitates efficient secretion of insulin-like peptides, not their storage in peroxisomes. Therefore, antibody-based detection of Dilp2 in peroxisomes would not test the hypothesis we put forward. Second, there are no peroxisomes present in pex19 or pex3 mutant backgrounds, so it is technically impossible to assess Golgi-peroxisome contacts or Dilp2 localization to peroxisomes in these animals. This limitation applies to all experiments aiming to visualize peroxisome-associated phenomena in these genetic contexts. Therefore, our current data integrate both structural and functional approaches to support the role of Golgi-peroxisome proximity in peptide secretion, while explicitly leaving room for future work to dissect mechanistic detail.

    2. We thank the reviewer for raising the important point of reagent validation. The RNAi lines used in this study were obtained from established community resources (VDRC, BDSC) where design minimizes off-target effects and in several cases, we employed more than one independent RNAi line or combined RNAi with mutant alleles to confirm phenotypes. For antibodies, all used reagents are either validated in previous publications (Li et al., 2015 Nature) or functionally validated in this work by showing loss of signal upon knockdown or mutation. For example, the antibody against PMP70 showed specific loss of staining in respective mutant backgrounds (Line 265), providing strong evidence of reagent specificity and efficacy. Importantly, the consistency of our findings across independent reagents provides confidence that the reported phenotypes reflect specific gene functions rather than off-target effects.

    3. We thank the reviewers for raising this point. The behavior of control animals under nutritional manipulation has been well characterized by Bisen et al. (eLife 2024), who show that IPC activity decreases upon starvation and normalizes upon refeeding. Our results are fully consistent with these published controls, and we have cited this work in the Results section (Line 182). Importantly, our novel finding is that pex19 mutants maintain this nutritional state-dependent modulation of IPC activity but nevertheless fail to release neuropeptides, thereby uncoupling activity from secretion. This extends the published control framework and provides new mechanistic insight into the role of peroxisomes in peptide secretion.

    Minor Points

    1. We will revise the figure legend to explicitly state that the image shows Dilp2 antibody staining, not a transgene.

    2. We will expand the schematic and legend to clarify the sequence of peroxisome biogenesis factors: Pex19/3/16 act early during membrane formation, whereas Pex2/1 function in matrix import.

    3. We appreciate the reviewers insightful point regarding the route of Dilp2 secretion. The assay we employed is an established and widely used readout for insulin release in Drosophila (Ziegler A. et al., Scientific Reports 2018). This approach provides a robust and reproducible measure of relative secretion dynamics, since depletion of somatic Dilp2 stores strongly correlates with regulated release. We fully agree that secretion of Dilp2 physiologically occurs predominantly from axon terminals into the hemolymph, rather than directly from the soma. The somatic signal therefore reflects the mobilization and trafficking of the peptide toward secretory sites, and its loss has been consistently interpreted in the field as a proxy for secretion. To further strengthen our study, in a revised version of the manuscript we will include complementary data demonstrating Dilp2 levels in the hemolymph.

    4. We will correct the citation to refer to Figure 1C instead of Figure 2C.

    5. We thank the reviewer for this observation. The more robust rescue of survival with hemocyte-specific (hml-Gal4) Pex19 reconstitution compared to IPC-specific rescue likely reflects the broader systemic roles of hemocytes in fly physiology. Hemocytes not only participate in immune responses but also regulate metabolism, clear damaged cells, and secrete signaling molecules (Hersperger F. et al., eLife 2023, Bland M.L. Semin Cell Dev Biol. 2023). Restoring Pex19 in these cells can alleviate systemic metabolic imbalances, reduce toxic metabolites such as ROS and free fatty acids and improve overall survival. In contrast, IPC-specific rescue primarily restores Dilp2 secretion, addressing neuroendocrine defects but having limited effect on widespread metabolic dysfunction. It is, in fact, quite surprising that the IPC rescue alone leads to significant improvement in survival. This suggests a systemic effect of restoring peroxisome function only in these neuroendocrine cells. Our data imply that targeted restoration of peroxisome activity within IPCs can exert indirect whole-organism effects, possibly through improved insulin signaling, which is central to metabolic regulation. Thus, the stronger survival benefit from hemocyte rescue likely arises from non-cell-autonomous, systemic improvements beyond the neuroendocrine axis, while the IPC rescue result points to interesting global physiological consequences from correcting this pathway merely in IPCs.

    6. We thank the reviewer for the suggestion to clarify terminology. By “depletion of peroxisomes”, we refer to loss of functional peroxisomal membranes and failure to import matrix proteins, both essential for peroxisome biogenesis. Matrix proteins are soluble luminal enzymes imported via PTS1/PTS2 signals. “Ghosts” are membrane remnants lacking matrix proteins, seen when matrix import factors (e.g., Pex2) are disrupted, whereas loss of membrane biogenesis factors (Pex3/16/19) eliminates detectable peroxisomal membranes entirely. This distinction is key, as disruption of membrane factors impairs Dilp2 secretion, while matrix import defects do not, consistent with the residual “ghosts” supporting partial function. We will revise the text to define these terms clearly.

    7. We will remove the redundant panel from the supplement to avoid confusion.

    8. We thank the reviewer for raising this point. BODIPY-Ceramide is a well-established fluorescent shingolipid analog whose localization is dynamic and context-dependent. It is initially taken up and accumulates in the Golgi, supporting its common use as a Golgi marker (Pagano et al., 1991). Over time, it can traffic to other membranes, including the plasma membrane and endosomes (Marks D.L. 2008). Thus, Golgi labeling reflects early uptake while plasma membrane labeling corresponds to later trafficking. We will clarify this dynamic distribution in the revised text. Furthermore, we will provide an unsaturated version of Figure 4E in the supplement.

    9. We thank the reviewer for raising this point. We will test Dilp2 release in px-mCherry and other Golgi marker backgrounds to ensure these tools do not perturb Golgi-peroxisome interactions. We will include these data in the supplement.

    10. We will implement a consistent color scheme across all panels (Golgi vs. peroxisomes) and add individual legends for clarity.

    11. We will quantify Golgi size across feeding states. If significant changes are observed, we will include them in the manuscript; if not, we will state that explicitly and provide the quantification in the supplement.

    12. We will add a brief explanation: “BioGRID is an open-source database of protein and genetic interactions; we used it to identify potential Golgi-peroxisome contact regulators.”

    13. We thank the reviewer for raising this point. GM130 is not a luminal peroxisomal matrix protein but a peripheral membrane protein bound to the cytoplasmic face of the cis-Golgi (Nakamura et al., 1995, Barr et al., 1998). Our use of “matrix protein” referred to its structural role within the Golgi matrix scaffold, not to soluble luminal proteins. The accumulation observed in pex19 mutants therefore reflects mislocalization of a Golgi-associated protein rather than a peroxisomal matrix protein. To avoid confusion, we will revise the text to refer to GM130 as a “Golgi scaffold” or “Golgi peripheral membrane protein”.

    14. We thank the reviewer for this important point. While DCVs are classically reported as ~100-150 nm in diameter, our use of a 500 nm threshold reflects both technical and biological considerations. Confocal resolution (~200-300 nm laterally, worse axially) limits reliable detection of smaller vesicles and vesicle clusters often appear as single, larger objects after 3D segmentation. In addition, DCV size is not fixed but varies with cell type, developmental stage and cargo load, with larger vesicles and aggregates reported in neuroendocrine cells (Kirchner et al., 2022). Our empirical threshold therefore enabled robust and reproducible identification of vesicle-like structures while minimizing background. Importantly, the objects detected showed appropriate colocalization with secretory markers and correlated with functional readouts, supporting their biological relevance. We will revise the Methods to clarify this rationale.

    15. We will revise the figure legends and text to clarify that Figure 1E illustrates peroxisome maturation (contextual to Dilp2 biology rather than showing Dilp2 directly) and that Figure 6G highlights the relevant pathway.

    Disclosure: 

    Portions of my responses to the reviewer’s major and minor comments were refined with the assistance of OpenAI’s ChatGPT to improve clarity and tone.

    Competing interests

    The author of this comment declares that they have no competing interests.