Ceramide transfer protein regulates G-protein coupled phospholipase signalling in Drosophila photoreceptors

  1. Shirish Mishra
  2. Ali Khan
  3. Vaisaly R Nath
  4. Tejaswini Manoj
  5. BG Thejaswini
  6. Amruta Naik
  7. Ujjaini Dasgupta
  8. Padinjat Raghu

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

The non-vesicular transfer of lipids between organelles at membrane contact sites (MCS) has been proposed as a key principle in the regulation of cell physiology. While several proteins with lipid transfer activity have been identified and localized to MCS, their functional significance for supporting physiology is poorly understood. Ceramide transfer protein (CERT) is one such molecule that can transfer ceramide between membranes in vitro . However, evidence for the mechanism and in vivo significance of CERT function is limited. In this study, we have analyzed the function of the only gene ( dcert ) encoding CERT in Drosophila . We find that loss of function alleles of dcert ( dcert 1 ), show elevated levels of short chain ceramide species along with a reduction in the levels of its metabolite phosphatidyl ethanolamine ceramide. Physiological analysis of dcert 1 mutant alleles showed reduced electrical responses in the eye to light stimulation although photoreceptors did not undergo retinal degeneration, and this phenotype could be rescued by reconstitution of dcert 1 with the wild type gene. The altered light response in dcert 1 was associated with a reduction in the rate of phosphatidylinositol 4,5 bisphosphate (PIP 2 ) resynthesis following light induced phospholipase C (PLC) stimulation. The reduced electrical response of dcert 1 could be suppressed by reducing ceramide synthesis at the ER. Taken together, our findings suggest that ceramide synthesized at the ER and transferred to the Golgi by CERT regulates G-protein coupled phospholipase C signaling in vivo .

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

    Manuscript number: RC-2025-03160

    Corresponding author(s) Padinjat, Raghu

    [The “revision plan” should delineate the revisions that authors intend to carry out in response to the points raised by the referees. It also provides the authors with the opportunity to explain their view of the paper and of the referee reports.

    The document is important for the editors of affiliate journals when they make a first decision on the transferred manuscript. It will also be useful to readers of the reprint and help them to obtain a balanced view of the paper.

    If you wish to submit a full revision, please use our "Full Revision" template. It is important to use the appropriate template to clearly inform the editors of your intentions.]

    1. General Statements [optional]

    We thank all three reviewers for appreciating the novelty of our analysis of CERT function in a physiological context in vivo. While many studies have been published on the biochemistry and function of CERT in cultured cells, there are limited studies, if any, relating the impact of CRT function at the biochemical level to its function on a physiological process, in our case the electrical response to light.

    We also that all reviewers for commenting on the importance of our rescue of dcert mutants with hCERT and the scientific insights raised by this experiment. All reviewers have also noted the importance of strengthening our observation that hCERT, in these cells, is localized at ER-PM MCS rather that the more widely reported localization at the Golgi. We highlight that many excellent studies which have localized CERT at the Golgi are performed in cultured, immortalized, mammalian cells. There are limited studies on the localization of this protein in primary cells, neurons or in polarized cells. With the additional experiments we have proposed in the revision for this aspect of the manuscript, we believe the findings will be of great novelty and widespread interest.

    We believe we can address almost all points raised by reviewers thereby strengthening this exciting manuscript.

    2. Description of the planned revisions

    Insert here a point-by-point reply that explains what revisions, additional experimentations and analyses are planned to address the points raised by the referees.

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

    This manuscript dissects the physiological function of ceramide transfer protein (CERT) by studying the phenotype of CERT null Drosophila.

    dCERT null animals have a reduced electrical response to light in their photoreceptors, reduced baseline PIP2 accumulation in the cells and delayed re-synthesis of PIP2 and its precursor, PI4P after light stimulation. There are also reduced ER:PM contact sites at the rhabdomere and a corresponding reduction in the localization of PI/PA exchange protein, RDGB at this site. Therefore, the animals seem to have an impaired ability for sustaining phototransduction, which is nonetheless milder than that seen after loss of RDGB, for example. In terms of biochemical function, there is no overall change in ceramides, with some minor increases in specific short chain pools. There is however a large decrease in PE-ceramide species, again selective for a few molecular species. Curiously, decreasing ceramides with a mutant in ceramide synthesis is able to partially rescue both the electrical response and RDGB localization in dCERT flies, implying the increased ceramide species contribute to the phenotype. In addition, a mutation in PE-ceramide synthase largely phenocopies the dCERT null, exhiniting both increases ceramides and decreased PE-ceramide.

    In addition, dCERT flies were shown to have reduced localization of some plasma membrane proteins to detergent-resistant membrane fractions, as well as up regulation of the IRE1 and PERK stress-response pathways. Finally, dCERT nulls could be rescued with the human CERT protein, demonstrating conservation of core physiological function between these animals. Surprisingly, CERT is reported to localize to the ER:PM junctions at rhabdomeres, as opposed to the expected ER:Golgi contact sites. Specific areas where the manuscript could be strengthened include:

    Figure 2 studies the phototransduction system. Although clear changes in PI4P and PIP2 are seen, it would be interesting to see if changed PA accumulation occur in the dCERT animals, since RDGB localization is disrupted: this is expected to cause PM PA accumulation along with reduced PIP2 synthesis.

    It is an important question raised by the reviewer to check PA levels. In the present study we have noticed that localization of RDGB at the base of the rhabdomere in dcert1 is reduced but not completely removed. Consequently, one may consider the situation on dcert1 as a partial loss of function of RDGB and consistent with this, the delay in PI4P and PI(4,5)P2 resynthesis is not as severe as in rdgB9 which is a strong hypomorph (PMID: 26203165).

    rdgB9 mutants also show an elevation in PA levels and the reviewer is right that one might expect changes in PA levels too as RDGB is a PI/PA transfer protein. We expect that if measured, there will be a modest elevation in PA levels. However, previous work has shown that elevation of PA levels at the or close to the rhabdomere lead to retinal degeneration Specifically, elevated PA levels by dPLD overexpression disrupts rhabdomere biogenesis and leads to retinal degeneration (PMID: 19349583). Similarly, loss of the lipid transfer protein RDGB leads to photoreceptor degeneration (PMID: 26203165). In this study, we report that retinal degeneration is not a phenotype of* dcert1*. Thus measurements of PA levels though interesting may not be that informative in the context of the present study. However, if necessary, we can measure PA levels in dcert1.

    Lines 228-230 state: "These findings suggest an important contribution for reduced PE - Cer levels in the eye phenotypes of dcert". Does it not also suggest a contribution of the elevated ceramide species, since these are also observed in the CPES animals?

    We agree with the reviewer that not only reduced PE-Ceramide but also elevated ceramide levels in GMR>CPESi could contribute to the eye phenotype. This statement will be revised to reflect this conclusion.

    Figure 6D is a key finding that human CERT localized to the rhabdomere at ER:PM contact sites, though the reviewer was not convinced by these images. Is the protein truly localized to the contact sites, or simply have a pool of over-expressed protein localized to the surrounding cytoplasm? It also does not rule out localization (and therefore function) at ER:PM contact sites.

    Since hCERT completely rescued eye phenotype of dcert1 the localization we observe for hCERT must be at least partly relevant. We will perform additional IHC experiments to

    • Co-localize hCERT with an ER-PM MCS marker, e.g RDGB in wild type flies
    • Co-localize hCERT with VAP-A that is enriched at the ER-PM MCS. This should help to determine if there are MCS and non-MCS pools of hCERT in these cells. marker, e.g RDGB in wild type flies
    • Test if there is a pool of hCERT, in these cells that also localizes (or not) with the Golgi marker Golgin 84. These will be included in the revision to strengthen this important point.

    Statistics: There are a large number of t-tests employed that do not correct for multiple comparisons, for example in figures 3B, 3D, 3H, 4C, 6C, S2A, S2B, S3B and S3C.

    We will performed multiple comparisons with mentioned data and incorporate in the revised manuscript.

    There are two Western blotting sections in the methods.

    The first Western blotting methods is for general blots in the paper. The second western blotting section is related to the samples from detergent resistant membrane (DRM) fractions. We will clearly explain this information in the methods section of the manuscript.

    Reviewer #1 (Significance (Required)):

    Overall, the manuscript is clearly and succinctly written, with the data well presented and mostly convincing. The paper demonstrates clear phenotypes associated with loss of dCERT function, with surprising consequences for the function of a signaling system localized to ER:PM contact sites. To this reviewer, there seem to be three cogent observations of the paper: (i) loss of dCERT leads to accumulation of ceramides and loss of PE-ceramide, which together drive the phenotype. (ii) this ceramide alteration disrupts ER:PM contact sites and thus impairs phototransduction and (iii) rescue by human CERT and its apparent localization to ER:PM contact sites implies a potential novel site of action. Although surprising and novel, the significance of these observations are a little unclear: there is no obvious mechanism by which the elevated ceramide species and decreased PE-ceramide causes the specific failure in phototrasnduction, and the evidence for a novel site of action of CERT at the ER:PM contact sites is not compelling. Therefore, although an interesting and novel set of observations, the manuscript does not reveal a clear mechanistic basis for CERT physiological function.

    We thank reviewer for appreciating the quality of our manuscript while also highlighting points through which its impact can be enhanced. To our knowledge this is one of the first studies to tackle the challenging problem of a role for CERT in physiological function. We would like to highlight two points raised:

    • We do understand that the localisation of hCERT at ER-PM MCS is unusual compared to the traditional reported localization to ER-Golgi sites. This is important for the overall interpretation of the results in the paper on how dCERT regulates phototransduction. As indicated in response to an earlier comment by the reviewer we will perform additional experiments to strengthen our conclusion of the localization of hCERT.
    • With regard to how loss of dCERT affects phototransduction, we feel to likely mechanisms contribute. If the localization of hCERT to ER-PM MCS is verified through additional experiments (see proposal above) then it is important to note that ER-PM MCS in these cells includes the SMC (smooth endoplasmic reticulum) the major site of lipid synthesis. It is possible that loss of dCERT leads to ceramide accumulation in the smooth ER and disruption of ER-PM contacts. That may explain why reducing the levels of ceramide at this site partially rescues the eye phenotype.

    The multi-protein INAD-TRP-NORPA complex, central to phototransduction have previously been shown to localise to DRMs in photoreceptors. PE-Ceramides are important contributors to the formation of plasma membrane DRMs and we have presented biochemical evidence that the formation of these DRMs are reduced in the dcert1. This may be a mechanism contributing to reduced phototransduction. This latter mechanism has been proposed as a physiological function of DRMs but we think our data may be the first to show it in a physiological model.

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

    Summary Non-vesicular lipid transfer by lipid transfer proteins regulates organelle lipid compositions and functions. CERT transfers ceramide from the ER to Golgi to produce sphingomyelin, although CERT function in animal development and physiology is less clear. Using dcert1 (a protein-null allele), this paper shows a disruption of the sole Drosophila CERT gene causes reduced ERG amplitude in photoreceptors. While the level and localization of phototransduction machinery appears unaffected, the level of PIP2 and the localization of RDGB are perturbed. Collectively, these observations establish a novel link between CERT and phospholipase signaling in phototransduction. To understand the molecular mechanism further, the authors performed lipid chromatography and mass spec to characterize ceramide species in dcert1. This analysis reveals that whereas the total ceramide remains unaffected, most PE-ceramide species are reduced. The authors use lace mutant (serine palmitoyl transferase) and CPES (ceramide phosphoethanolamine synthase) RNAi to distinguish whether it is the accumulation of ceramide in the ER or the reduction of sphingolipid derivates in the Golgi that is the cause for the reduced ERG amplitude. Mutating one copy of lace reduces ceramide level by 50% and partially rescues the ERG defect, suggesting that the accumulation of ceramide in the ER is a cause. CPES RNAi phenocopies the reduced ERG amplitude, suggesting the production of certain sphingolipid is also relevant.

    Major comments:

    1. By showing the reduced PIP2 level, the decreased SMC sites at the base of rhabdomeres, and the diffused RDGB localization in dcert1, the authors favor the model, in which the disruption of ceramide metabolism affects PIP transport. However, it is unclear if the reduced PIP2 level (i.e., reduced PH-PLCd::GFP staining) is specific to the rhabdomeres. It should be possible to compare PH-PLCd::GFP signals in different plasma membranes between wildtype and dcert1. If PH-PLCd::GFP signal is specifically reduced at the rhabdomeres, this conclusion will be greatly strengthened. In addition, the photoreceptor apical plasma membrane includes rhabdomere and stalk membrane. Is the PH-PLCd::GFP signal at the stalk membrane also affected?

    Due to the physical organization of optics in the fly eye, the pseudopupil imaging method used in this study collects the signal for the PIP2 probe (PH-PLCd::GFP) mainly from the apical rhabdomere membrane of photoreceptors in live imaging experimental mode. Therefore, the PIP2 signal from these experiments cannot be used to interpret the level of PIP2 either at the stalk membrane or indeed the basolateral membrane.

    The point raised by the reviewer, i.e whether CERT selectively controls PIP2 levels at the rhabdomere membrane or not, is an interesting one. To do this, we will need to fix fly photoreceptors and determine the PH-PLCd::GFP signal using single slice confocal imaging. When combined with a stalk marker such as CRUMBS, it should be possible to address the question of which are the membrane domains at which dCERT controls PIP2 levels. If the sole mechanism of action of dCERT is via disruption of ER-PM MCS then only the apical rhabdomere membrane PIP2 should be affected leaving the stalk membrane and basolateral membrane unaffected.

    Thank you very much for raising this specific point.

    The analysis of RDGB localization should be done in mosaic dcert1 retinas, which will be more convincing with internal control for each comparison. In addition, the phalloidin staining in Figure 2J shows distinct patterns of adherens junctions, indicating that the wildtype and dcert1 were imaged at different focal planes.

    We understand that having mosaics is an alternative an elegant way to perform a a side by side analysis of control and mutant. However this would require significant investment of time and effort, perhaps beyond the scope of this study. If we were to perform a mosaic analysis, this would compromise our ERG analysis since ERG is an extracellular recording We feel that this is beyond the scope of this study and perhaps may not be necessary as such (see below).

    In the revision we will present equivalent sections of control and dcert1 taken from the nuclear plane of the photoreceptor. This should resolve the reviewer’s concerns.

    The significance of ceramide species levels in dcert1 and GMR>CPESRNAi needs to be explained better. Do certain alterations represent accumulation of ceramides in the ER?

    Species level analysis of changes in ceramides reveal that elevations in *dcert1 *are seen mainly in the short chain ceramides (14 and 16 carbon chains). These most likely represent the short chain ceramides synthesised in the ER and accumulating due to the block in further metabolism to PE-Cer due to depletion in CPES.

    Species level analysis of changes in ceramides reveal that in dcert1 there is a ceramide transport related defect leading to elevation, primarily, in the short chain ceramides (14 and 16 carbon chains), and this selective supply defect leads to a reduction in PE-Cer levels, with a maximum change in the ratio of short-chain Cer:PE Cer (Figure 3A-D). Though there is no apparent change in the total ceramide level the species specific elevation in the ceramides disturb the fine -balance between the short-chain ceramides and the long and very-long chain ceramides. As the function of long and very-long chain ceramides are implicated in dendrite development and neuronal morphology (doi: 10.1371/journal.pgen.1011880), therefore this alteration in the fine balance between different ceramide species probably impacts the integrity and fluidity of the membrane environment. On the other hand it leads to a possibility of a defined function of the short-chain ceramides in electrical responses to light signalling in the eye, especially with respect to the PE-ceramides that are reduced by around 50%.

    In contrast the GMR>CPESRNAi leads to more of a substrate accumulation showing ceramide increase (14, 16, 18, 20 carbon chains) and decrease in PE-Cer levels (Figure 4D, E). In this case Cer accumulation is due to the block in further metabolism to PE-Cer arising from depletion in CPES.

    We will include this in the discussion of a revised version.

    The suppression by lace is interpreted as evidence that the reduced ERG amplitude in dcert1 is caused by ceramide accumulation in the ER. This interpretation seems preliminary as lace may interact with dcert genetically by other mechanisms.

    The dcert1 mutant exhibits increased levels of short-chain ceramides (Fig 3B), whereas the lace heterozygous mutant (laceK05305/+) displays reduced short-chain ceramide levels (Supp Fig 2B). In the laceK05305/+; dcert1 double mutant, ceramide levels are lower than those observed in the dcert1 mutant alone (Supp Fig 2B), indicating a partial genetic rescue of the elevated ceramide phenotype.

    Furthermore, through multiple independent genetic manipulations that modulate ceramide metabolism (alterations of dcert, cpes and lace), we consistently observe that increased ceramide levels correlate with a reduction in ERG amplitude, suggesting that ceramide accumulation negatively impacts photoreceptor function. Taken together, these observations indicate that the reduction in ceramide levels in the laceK05305/+; dcert1 double mutant likely contributes to the suppression of the ERG defect observed in the dcert1 mutant.

    The authors show that ERG amplitude is reduced in GMR>CPESRNAi. While this phenocopying is consistent with the reduced ERG amplitude in dcert1 being caused by reduced production of PE-ceramide, GMR>CPESRNAi also shows an increase in total ceramide level. Could this support the hypothesis that reduced ERG amplitude is caused by an accumulation of ceramide elsewhere? In addition, is the ERG amplitude reduction in GMR>CPESRNAi sensitive to lace?

    We agree that in addition to reduced PE-Ceramide, the elevated ceramide levels in GMR>CPESi could contribute to the eye phenotype. We will introduce lace heterozygous mutant in the *GMR>CPESi *background to test the contribution of elevated ceramide levels in the *GMR>CPESi * background and incorporate the data in the revision. Thank you for this suggestion.

    Along the same line, while the total ceramide level is significantly reduced in lace heterozygotes, is the PE-ceramide level also reduced? If yes, wouldn't this be contradictory to PE-ceramide production being important for ERG amplitude?

    Mass spec measurements show that levels of PE-Cer were not reduced in lacek05305/+ compared to wild type. This data will be included in the revised manuscript. However, the ERG amplitude of these flies and also in those with lace depletion using two independent RNAi lines were not reduced.

    What is the explanation and significance for the age-dependent deterioration of ERG amplitude in dcert1? Likewise, the significance of no retinal degeneration is not clearly presented.

    There could be multiple reasons for the age dependent deterioration of the ERG amplitude, in the absence of retinal degeneration. Drosophila phototransduction cascade depends heavily on ATP production. The age dependent reduction in ATP synthesis could lead to deterioration in the ERG amplitude. These may include instability of the DRMs due to reduced PE-Cer, lower ATP levels due to mitochondrial dysfunction, an perhaps others. A previous study has shown that ATP production is highly reduced along with oxidative stress and metabolic dysfunction in dcert1 flies aged to 10 days and beyond (PMID: 17592126). The same study has also found no neuronal degeneration in dcert1 that phenocopies absence of photoreceptor degeneration in the present study. We will attempt a few experiments to rule in or rule out the these and revise the discussion accordingly.

    The rescue of dcert1 phenotype by the expression of human CERT is a nice result. In addition to demonstrating a functional conservation, it allows a determination of CERT protein localization. However, the quality of images in Figure 6D should be improved. The phalloidin staining was rather poor, and the CNX99A in the lower panel was over-exposed, generating bleed-through signals at the rhabdomeres. In addition, the localization of hCERT should be explored further. For instance, does hCERT colocalize with RDGB? Is the hCERT localization altered in lace or GMR>CPESRNAi background?

    As indicated in response to reviewer 1:

    We will perform additional IHC experiments to

    • Co-localize hCERT with an ER-PM MCS marker, e.g RDGB in wild type flies
    • Co-localize hCERT with VAP-A that is enriched at the ER-PM MCS. This should help to determine if there are MCS and non-MCS pools of hCERT in these cells. marker, e.g RDGB in wild type flies
    • Test if there is a pool of hCERT, in these cells that also localizes (or not) with the Golgi marker Golgin 84. These will be included in the revision to strengthen this important point.

    We will also attempt to perform hCERT localization in lace or GMR>CPESRNAi background

    Minor comments:

    1. In Line 128, Df(732) should be Df(3L)BSC732.

    Changes will be incorporated in the main manuscript.

    GMR-SMSrRNAi shows an increase in ERG peak amplitude. Is there an explanation for this?

    GMR-SMSrRNAi did show slight increase in ERG peak amplitude but was not statistically significant.

    Reviewer #2 (Significance (Required)):

    Significance As CERT mutations are implicated in human learning disability, a better understanding of CERT function in neuronal cells is certainly of interest. While the link between ceramide transport and phospholipase signaling is novel and interesting, this paper does not clearly explain the mechanism. In addition, as the ERG were measured long after the retinal cells were deficient in CERT or CPES, it is difficult to assess whether the observed phenotype is a primary defect. Furthermore, the quality of some images needs to be improved. Thus, I feel the manuscript in its current form is too preliminary.

    We thank reviewer for highlighting the importance and significance of our work in the light of recent studies of CERT function in ID. As with all genetic studies it is difficult to completely disentangle the role of a gene during development from a role only in the adult. However, we will attempt to perhaps use the GAL80ts system to uncouple these two potential components of CERT function in photoreceptors. The goal will be to determine if CERT has a specific role only in adult photoreceptors or if this is coupled to a developmental role. Since ID is as a neurodevelopmental disorder, a developmental role for CERT would be equally interesting.

    As previously indicated images will be improved bearing in mind the reviewer comments.

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

    Summary: Lipid transfer proteins (LTPs) shuttle lipids between organelle membranes at membrane contact sites (MCSs). While extensive biochemical and cell culture studies have elucidated many aspects of LTP function, their in vivo physiological roles are only beginning to be understood. In this manuscript, the authors investigate the physiological role of the ceramide transfer protein (CERT) in Drosophila adult photoreceptors-a model previously employed by this group to study LTP function at ER-PM contact sites under physiological conditions. Using a combination of genetic, biochemical, and physiological approaches, they analyze a protein-null mutant of dcert. They show that loss of dcert causes a reduction in electrical response to light with progressive decrease in electroretinogram (ERG) amplitude with age but no retinal degeneration. Lipidomic analysis shows that while the total levels of ceramides are not changed in dcert mutants, they do observe significant change in certain species of ceramides and depletion of downstream metabolite phosphoethanolamine ceramide (PE-Cer). Using fluorescent biosensors, the authors demonstrate reduced PIP2 levels at the plasma membrane, unchanged basal PI4P levels and slower resynthesis kinetics of both lipids following depletion. Electron microscopy and immunolabeling further reveal a reduced density of ER-PM MCSs and mislocalization of the MCS-resident lipid transfer protein RDGB. Genetic interaction studies with lace and RNAi-mediated knockdown of CPES support the conclusion that both ER ceramide accumulation and PM PE-Cer depletion contribute to the observed defects in dcert mutants. In addition, detergent-resistant membrane fractionation indicates altered plasma membrane organization in the absence of dcert. The study also reports upregulation of unfolded protein response transcripts, including IRE1 and PERK, suggesting increased ER stress. Finally, expression of human CERT rescues the reduced electrical response, demonstrating functional conservation across species. Overall the manuscript is well written that builds on established work and experiments are technically rigorous. The results are clearly presented and provide valuable insights into the physiological role of CERT.

    Major comments: 1.The reduced ERG amplitude appears to be the central phenotype associated with the loss of dcert, and most of the experiments in this manuscript effectively build a mechanistic framework to explain this observation. However, the experiments addressing detergent-resistant membrane domains (DRMs) and the unfolded protein response (UPR) seem somewhat disconnected from the main focus of the study. The DRM and UPR data feel peripheral and could benefit from few experiments for functional linkage to the ERG defect or should be moved to supplementary.

    We agree with the reviewer that further experiments are needed to link the DRM data to the ERG defects. That would need specific biochemical alteration at the PM to modulate PE-Cer species and their effect on scaffolding proteins required for phototransduction (that is beyond the scope of the present study). We will consider moving these to the supplementary section as suggested by the reviewer.

    2.The changes in ceramide species and reduction in PE-Cer are key findings of the study. These results should be further validated by performing a genetic rescue using the BAC or hCERT fly line to confirm that the lipidomic changes are specifically due to loss of CERT function.

    Thank you for this comment. We will include this in the revised manuscript.

    3.Figure 2B-C and 2E-F: Representative images corresponding to the quantified data should be included to illustrate the changes in PIP2 and PI4P reporters. Given that the fluorescence intensity of the PIP2 reporter at the PM is reduced in the dcert mutant relative to control, the authors should also verify that the reporter is expressed at comparable levels across genotypes.

    • As mentioned by the reviewer we will include representative images alongside our quantified data both of the basal ones and that of the kinetic study.

    • Western blot of reporters (PH-PLCd::GFP and P4M::GFP) across genotypes will be added to the revised manuscript. 4.Figure 2J-K: The partial mislocalization of RDGB represents an important observation that could mechanistically explain the reduced resynthesis of PI4P and PIP2 and consequently, the decreased ERG amplitude in dcert mutants. However, this result requires further validation. First, the authors should confirm whether this mislocalization is specific to RDGB by performing co-staining with another ER-PM MCS marker, such as VAP-A, to assess whether overall MCS organization is disrupted. Second, the quantification of RDGB enrichment at ER-PM MCSs should be refined. From the representative images, RDGB appears redistributed toward the photoreceptor cell body, but the presented quantification does not clearly reflect this shift. The authors should therefore include an analysis comparing RDGB levels in the cell body versus the submicrovillar region across genotypes. This analysis should be repeated for similar experiments across the study. Additionally, the total RDGB protein level should be quantified and reported. Finally, since RDGB mislocalization could directly contribute to the decreased ERG amplitude, it would be valuable to test whether overexpression of RDGB in dcert mutants can rescue the ERG phenotype.

    • In our ultrastructural studies (Fig. 2H, 2I and Sup. Fig. 1A, 1B) we did see reduction in PM-SMC MCS that was corroborated with RDGB staining.

    • Comparative ratio analysis of RDGB localisation at ER-PM MCS vs cell body will be included in the manuscript for all RDGB staining.

    • We have done western analysis for total RDGB protein level in ROR and dcert1. This data will be included in the revised manuscript.

    • This is a very interesting suggestion and we will test if RDGB overexpression can rescue ERG phenotype in dcert1.

    5.Figure 3F and I-J: Inclusion of appropriate WT and laceK05205/+ controls is necessary to allow proper interpretation of the results. These controls would strengthen the conclusions regarding the functional relationship between dcert and lace.

    Changes will be incorporated as per the suggestion.

    6.Figure 5C: The representative images shown here appear to contradict the findings described in Figure 2A. In Figure 5C, Rhodopsin 1 levels seem markedly reduced in the dcert mutants, whereas the text states that Rh1 levels are comparable between control and mutant photoreceptors. The authors should replace or reverify the representative images to ensure that they accurately reflect the conclusions presented in the text.

    We will reverify the representative images and changes will be accordingly incorporated.

    7.Figure 6D: The reported localization of hCERT to ER-PM MCSs is a key and potentially insightful observation, as it suggests the subcellular site of dcert activity in photoreceptors. However, the representative images provided are not sufficiently conclusive to support this claim. The authors should validate hCERT localization by co-staining with established markers like RDGB for ER-PM CNX99A for the ER and a Golgi marker since mammalian CERT is classically localized to ER-Golgi interfaces. Optionally, the authors could also quantify the relative distribution of hCERT among these compartments to provide a clearer assessment of its subcellular localization.

    As indicated in response to reviewer 1:

    We will perform additional IHC experiments to

    • Co-localize hCERT with an ER-PM MCS marker, e.g RDGB in wild type flies
    • Co-localize hCERT with VAP-A that is enriched at the ER-PM MCS. This should help to determine if there are MCS and non-MCS pools of hCERT in these cells. marker, e.g RDGB in wild type flies
    • Test if there is a pool of hCERT, in these cells that also localizes (or not) with the Golgi marker Golgin 84. These will be included in the revision to strengthen this important point.

    Minor comments: 1.In the first paragraph of introduction, authors should consider citing few of the key MCS literature.

    Additional literature will be included as per the suggestion.

    2.Line 132: data not shown is not acceptable. Authors should consider presenting the findings in the supplemental figure.

    Data will be added in supplement as per the suggestion.

    3.The authors should include a comprehensive table or Excel sheet summarizing all statistical analyses. This should include the sample size, type of statistical test used and exact p-values. Providing this information will improve the transparency, reproducibility and overall rigor of the study.

    We will provide all the statistical analyses in mentioned format as per the suggestion.

    4.The materials and methods section can be reorganized to include citation for flystocks which do not have stock number or RRIDs if the stocks were previously described but are not available from public repositories. They should expand on the details of various quantification methods used in the study. Finally including a section of Statistical analyses would further enhance transparency and reproducibility

    • Stock details will be added wherever missing as per the suggestion.
    • Statistical analyses section will be included in the material and methods. **Referee cross-commenting**

    1.I concur with Reviewer 1 regarding the need for more detailed reporting of statistical analyses.

    We will perform multiple comparisons with mentioned data and incorporate in the revised manuscript.

    2.I also agree with Reviewer 3 that the discussion should be expanded to address the age-dependent deterioration of ERG amplitude observed in the dcert mutants. This progressive decline could provide valuable insight into the long-term requirement of CERT function and signaling capacity at the photoreceptor membrane.

    Expanded discussion on the age dependent ERG amplitude decline will be incorporated in the discussion as per the suggestion.

    Reviewer #3 (Significance (Required)):

    This study explores the physiological function of CERT, a LTP localized at MCSs in Drosophila photoreceptors and uncovers a novel role in regulating plasma membrane PE-Cer levels and GPCR-mediated signaling. These findings significantly advances our understanding of how CERT-mediated lipid transport regulates G-protein coupled phospholipase C signaling in vivo. This work also highlights Drosophila photoreceptors as a powerful system to analyze the physiological significance of lipid-dependent signaling processes. This work will be of interest to researchers in neuronal cell biology, membrane dynamics and lipid signaling community. This review is based on my expertise in neuronal cell biology.

    We thank the reviewer for appreciating the significance of our work from a neuroscience perspective.

    3. Description of the revisions that have already been incorporated in the transferred manuscript

    Please insert a point-by-point reply describing the revisions that were already carried out and included in the transferred manuscript. If no revisions have been carried out yet, please leave this section empty.

    4. Description of analyses that authors prefer not to carry out

    Please include a point-by-point response explaining why some of the requested data or additional analyses might not be necessary or cannot be provided within the scope of a revision. This can be due to time or resource limitations or in case of disagreement about the necessity of such additional data given the scope of the study. Please leave empty if not applicable.

    We can address all reviewer points in the revision. However, we will not be able to perform a mosaic analysis of the impact of *dcert1 *mutant in the retina. We feel this is beyond the scope of this revision. In our response, we have highlighted how controls included in the revision offset the need for a mosaic analysis at this stage.

  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:

    Lipid transfer proteins (LTPs) shuttle lipids between organelle membranes at membrane contact sites (MCSs). While extensive biochemical and cell culture studies have elucidated many aspects of LTP function, their in vivo physiological roles are only beginning to be understood. In this manuscript, the authors investigate the physiological role of the ceramide transfer protein (CERT) in Drosophila adult photoreceptors-a model previously employed by this group to study LTP function at ER-PM contact sites under physiological conditions. Using a combination of genetic, biochemical, and physiological approaches, they analyze a protein-null mutant of dcert. They show that loss of dcert causes a reduction in electrical response to light with progressive decrease in electroretinogram (ERG) amplitude with age but no retinal degeneration. Lipidomic analysis shows that while the total levels of ceramides are not changed in dcert mutants, they do observe significant change in certain species of ceramides and depletion of downstream metabolite phosphoethanolamine ceramide (PE-Cer). Using fluorescent biosensors, the authors demonstrate reduced PIP2 levels at the plasma membrane, unchanged basal PI4P levels and slower resynthesis kinetics of both lipids following depletion. Electron microscopy and immunolabeling further reveal a reduced density of ER-PM MCSs and mislocalization of the MCS-resident lipid transfer protein RDGB. Genetic interaction studies with lace and RNAi-mediated knockdown of CPES support the conclusion that both ER ceramide accumulation and PM PE-Cer depletion contribute to the observed defects in dcert mutants. In addition, detergent-resistant membrane fractionation indicates altered plasma membrane organization in the absence of dcert. The study also reports upregulation of unfolded protein response transcripts, including IRE1 and PERK, suggesting increased ER stress. Finally, expression of human CERT rescues the reduced electrical response, demonstrating functional conservation across species.Overall the manuscript is well written that builds on established work and experiments are technically rigorous. The results are clearly presented and provide valuable insights into the physiological role of CERT.

    Major comments:

    1.The reduced ERG amplitude appears to be the central phenotype associated with the loss of dcert, and most of the experiments in this manuscript effectively build a mechanistic framework to explain this observation. However, the experiments addressing detergent-resistant membrane domains (DRMs) and the unfolded protein response (UPR) seem somewhat disconnected from the main focus of the study. The DRM and UPR data feel peripheral and could benefit from few experiments for functional linkage to the ERG defect or should be moved to supplementary. 2.The changes in ceramide species and reduction in PE-Cer are key findings of the study. These results should be further validated by performing a genetic rescue using the BAC or hCERT fly line to confirm that the lipidomic changes are specifically due to loss of CERT function. 3.Figure 2B-C and 2E-F: Representative images corresponding to the quantified data should be included to illustrate the changes in PIP2 and PI4P reporters. Given that the fluorescence intensity of the PIP2 reporter at the PM is reduced in the dcert mutant relative to control, the authors should also verify that the reporter is expressed at comparable levels across genotypes. 4.Figure 2J-K: The partial mislocalization of RDGB represents an important observation that could mechanistically explain the reduced resynthesis of PI4P and PIP2 and consequently, the decreased ERG amplitude in dcert mutants. However, this result requires further validation. First, the authors should confirm whether this mislocalization is specific to RDGB by performing co-staining with another ER-PM MCS marker, such as VAP-A, to assess whether overall MCS organization is disrupted. Second, the quantification of RDGB enrichment at ER-PM MCSs should be refined. From the representative images, RDGB appears redistributed toward the photoreceptor cell body, but the presented quantification does not clearly reflect this shift. The authors should therefore include an analysis comparing RDGB levels in the cell body versus the submicrovillar region across genotypes. This analysis should be repeated for similar experiments across the study. Additionally, the total RDGB protein level should be quantified and reported. Finally, since RDGB mislocalization could directly contribute to the decreased ERG amplitude, it would be valuable to test whether overexpression of RDGB in dcert mutants can rescue the ERG phenotype. 5.Figure 3F and I-J: Inclusion of appropriate WT and laceK05205/+ controls is necessary to allow proper interpretation of the results. These controls would strengthen the conclusions regarding the functional relationship between dcert and lace. 6.Figure 5C: The representative images shown here appear to contradict the findings described in Figure 2A. In Figure 5C, Rhodopsin 1 levels seem markedly reduced in the dcert mutants, whereas the text states that Rh1 levels are comparable between control and mutant photoreceptors. The authors should replace or reverify the representative images to ensure that they accurately reflect the conclusions presented in the text. 7.Figure 6D: The reported localization of hCERT to ER-PM MCSs is a key and potentially insightful observation, as it suggests the subcellular site of dcert activity in photoreceptors. However, the representative images provided are not sufficiently conclusive to support this claim. The authors should validate hCERT localization by co-staining with established markers like RDGB for ER-PM CNX99A for the ER and a Golgi marker since mammalian CERT is classically localized to ER-Golgi interfaces. Optionally, the authors could also quantify the relative distribution of hCERT among these compartments to provide a clearer assessment of its subcellular localization.

    Minor comments:

    1.In the first paragraph of introduction, authors should consider citing few of the key MCS literature. 2.Line 132: data not shown is not acceptable. Authors should consider presenting the findings in the supplemental figure. 3.The authors should include a comprehensive table or Excel sheet summarizing all statistical analyses. This should include the sample size, type of statistical test used and exact p-values. Providing this information will improve the transparency, reproducibility and overall rigor of the study. 4.The materials and methods section can be reorganized to include citation for flystocks which do not have stock number or RRIDs if the stocks were previously described but are not available from public repositories. They should expand on the details of various quantification methods used in the study. Finally including a section of Statistical analyses would further enhance transparency and reproducibility

    Referee cross-commenting

    1.I concur with Reviewer 1 regarding the need for more detailed reporting of statistical analyses. 2.I also agree with Reviewer 3 that the discussion should be expanded to address the age-dependent deterioration of ERG amplitude observed in the dcert mutants. This progressive decline could provide valuable insight into the long-term requirement of CERT function and signaling capacity at the photoreceptor membrane.

    Significance

    This study explores the physiological function of CERT, a LTP localized at MCSs in Drosophila photoreceptors and uncovers a novel role in regulating plasma membrane PE-Cer levels and GPCR-mediated signaling. These findings significantly advances our understanding of how CERT-mediated lipid transport regulates G-protein coupled phospholipase C signaling in vivo. This work also highlights Drosophila photoreceptors as a powerful system to analyze the physiological significance of lipid-dependent signaling processes. This work will be of interest to researchers in neuronal cell biology, membrane dynamics and lipid signaling community. This review is based on my expertise in neuronal cell biology.

  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

    Non-vesicular lipid transfer by lipid transfer proteins regulates organelle lipid compositions and functions. CERT transfers ceramide from the ER to Golgi to produce sphingomyelin, although CERT function in animal development and physiology is less clear. Using dcert1 (a protein-null allele), this paper shows a disruption of the sole Drosophila CERT gene causes reduced ERG amplitude in photoreceptors. While the level and localization of phototransduction machinery appears unaffected, the level of PIP2 and the localization of RDGB are perturbed. Collectively, these observations establish a novel link between CERT and phospholipase signaling in phototransduction. To understand the molecular mechanism further, the authors performed lipid chromatography and mass spec to characterize ceramide species in dcert1. This analysis reveals that whereas the total ceramide remains unaffected, most PE-ceramide species are reduced. The authors use lace mutant (serine palmitoyl transferase) and CPES (ceramide phosphoethanolamine synthase) RNAi to distinguish whether it is the accumulation of ceramide in the ER or the reduction of sphingolipid derivates in the Golgi that is the cause for the reduced ERG amplitude. Mutating one copy of lace reduces ceramide level by 50% and partially rescues the ERG defect, suggesting that the accumulation of ceramide in the ER is a cause. CPES RNAi phenocopies the reduced ERG amplitude, suggesting the production of certain sphingolipid is also relevant.

    Major comments:

    1. By showing the reduced PIP2 level, the decreased SMC sites at the base of rhabdomeres, and the diffused RDGB localization in dcert1, the authors favor the model, in which the disruption of ceramide metabolism affects PIP transport. However, it is unclear if the reduced PIP2 level (i.e., reduced PH-PLC::GFP staining) is specific to the rhabdomeres. It should be possible to compare PH-PLC::GFP signals in different plasma membranes between wildtype and dcert1. If PH-PLC::GFP signal is specifically reduced at the rhabdomeres, this conclusion will be greatly strengthened. In addition, the photoreceptor apical plasma membrane includes rhabdomere and stalk membrane. Is the PH-PLC::GFP signal at the stalk membrane also affected?
    2. The analysis of RDGB localization should be done in mosaic dcert1 retinas, which will be more convincing with internal control for each comparison. In addition, the phalloidin staining in Figure 2J shows distinct patterns of adherens junctions, indicating that the wildtype and dcert1 were imaged at different focal planes.
    3. The significance of ceramide species levels in dcert1 and GMR>CPESRNAi needs to be explained better. Do certain alterations represent accumulation of ceramides in the ER?
    4. The suppression by lace is interpreted as evidence that the reduced ERG amplitude in dcert1 is caused by ceramide accumulation in the ER. This interpretation seems preliminary as lace may interact with dcert genetically by other mechanisms.
    5. The authors show that ERG amplitude is reduced in GMR>CPESRNAi. While this phenocopying is consistent with the reduced ERG amplitude in dcert1 being caused by reduced production of PE-ceramide, GMR>CPESRNAi also shows an increase in total ceramide level. Could this support the hypothesis that reduced ERG amplitude is caused by an accumulation of ceramide elsewhere? In addition, is the ERG amplitude reduction in GMR>CPESRNAi sensitive to lace?
    6. Along the same line, while the total ceramide level is significantly reduced in lace heterozygotes, is the PE-ceramide level also reduced? If yes, wouldn't this be contradictory to PE-ceramide production being important for ERG amplitude?
    7. What is the explanation and significance for the age-dependent deterioration of ERG amplitude in dcert1? Likewise, the significance of no retinal degeneration is not clearly presented.
    8. The rescue of dcert1 phenotype by the expression of human CERT is a nice result. In addition to demonstrating a functional conservation, it allows a determination of CERT protein localization. However, the quality of images in Figure 6D should be improved. The phalloidin staining was rather poor, and the CNX99A in the lower panel was over-exposed, generating bleed-through signals at the rhabdomeres. In addition, the localization of hCERT should be explored further. For instance, does hCERT colocalize with RDGB? Is the hCERT localization altered in lace or GMR>CPESRNAi background?

    Minor comments:

    1. In Line 128, Df(732) should be Df(3L)BSC732.
    2. GMR-SMSrRNAi shows an increase in ERG peak amplitude. Is there an explanation for this?

    Significance

    As CERT mutations are implicated in human learning disability, a better understanding of CERT function in neuronal cells is certainly of interest. While the link between ceramide transport and phospholipase signaling is novel and interesting, this paper does not clearly explain the mechanism. In addition, as the ERG were measured long after the retinal cells were deficient in CERT or CPES, it is difficult to assess whether the observed phenotype is a primary defect. Furthermore, the quality of some images needs to be improved. Thus, I feel the manuscript in its current form is too preliminary.

  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

    This manuscript dissects the physiological function of ceramide transfer protein (CERT) by studying the phenotype of CERT null Drosophila.

    dCERT null animals have a reduced electrical response to light in their photoreceptors, reduced baseline PIP2 accumulation in the cells and delayed re-synthesis of PIP2 and its precursor, PI4P after light stimulation. There are also reduced ER:PM contact sites at the rhabdomere and a corresponding reduction in the localization of PI/PA exchange protein, RDGB at this site. Therefore, the animals seem to have an impaired ability for sustaining phototransduction, which is nonetheless milder than that seen after loss of RDGB, for example. In terms of biochemical function, there is no overall change in ceramides, with some minor increases in specific short chain pools. There is however a large decrease in PE-ceramide species, again selective for a few molecular species. Curiously, decreasing ceramides with a mutant in ceramide synthesis is able to partially rescue both the electrical response and RDGB localization in dCERT flies, implying the increased ceramide species contribute to the phenotype. In addition, a mutation in PE-ceramide synthase largely phenocopies the dCERT null, exhiniting both increases ceramides and decreased PE-ceramide.

    In addition, dCERT flies were shown to have reduced localization of some plasma membrane proteins to detergent-resistant membrane fractions, as well as up regulation of the IRE1 and PERK stress-response pathways. Finally, dCERT nulls could be rescued with the human CERT protein, demonstrating conservation of core physiological function between these animals. Surprisingly, CERT is reported to localize to the ER:PM junctions at rhabdomeres, as opposed to the expected ER:Golgi contact sites.

    Specific areas where the manuscript could be strengthened include:

    Figure 2 studies the phototransduction system. Although clear changes in PI4P and PIP2 are seen, it would be interesting to see if changed PA accumulation occur in the dCERT animals, since RDGB localization is disrupted: this is expected to cause PM PA accumulation along with reduced PIP2 synthesis.

    Lines 228-230 state: "These findings suggest an important contribution for reduced PE - Cer levels in the eye phenotypes of dcert". Does it not also suggest a contribution of the elevated ceramide species, since these are also observed in the CPES animals?

    Figure 6D is a key finding that human CERT localized to the rhabdomere at ER:PM contact sites, though the reviewer was not convinced by these images. Is the protein truly localized to the contact sites, or simply have a pool of over-expressed protein localized to the surrounding cytoplasm? It also does not rule out localization (and therefore function) at ER:PM contact sites.

    Statistics: There are a large number of t-tests employed that do not correct for multiple comparisons, for example in figures 3B, 3D, 3H, 4C, 6C, S2A, S2B, S3B and S3C.

    There are two Western blotting sections in the methods.

    Significance

    Overall, the manuscript is clearly and succinctly written, with the data well presented and mostly convincing. The paper demonstrates clear phenotypes associated with loss of dCERT function, with surprising consequences for the function of a signaling system localized to ER:PM contact sites. To this reviewer, there seem to be three cogent observations of the paper: (i) loss of dCERT leads to accumulation of ceramides and loss of PE-ceramide, which together drive the phenotype. (ii) this ceramide alteration disrupts ER:PM contact sites and thus impairs phototransduction and (iii) rescue by human CERT and its apparent localization to ER:PM contact sites implies a potential novel site of action. Although surprising and novel, the significance of these observations are a little unclear: there is no obvious mechanism by which the elevated ceramide species and decreased PE-ceramide causes the specific failure in phototrasnduction, and the evidence for a novel site of action of CERT at the ER:PM contact sites is not compelling. Therefore, although an interesting and novel set of observations, the manuscript does not reveal a clear mechanistic basis for CERT physiological function.

  5. Version published to 10.1101/2025.08.05.668821 on bioRxiv