Ciliary and non-ciliary functions of CEP104 in Xenopus
This article has been Reviewed by the following groups
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
- Evaluated articles (Review Commons)
Abstract
Cep104 is a conserved protein essential for centriole and cilia function, with mutations linked to Joubert Syndrome. We investigated its role in Xenopus embryonic development, revealing that Cep104 is crucial for neural tube closure (NTC) through regulating apical constriction. We show that the role of Cep104 in cilia and hedgehog signalling cannot alone explain the elicited defects. We go on to show that Cep104 localizes to the ends of cytoplasmic microtubules, influencing their stability. Downregulation of CEP104 led to microtubule instability and defects in multiciliated cell intercalation, a process dependent on stable microtubules. Our findings demonstrate that Cep104 functions beyond cilia, playing a significant role in cytoplasmic microtubule dynamics, suggesting that both ciliary and non-ciliary roles are important for neurodevelopment and the pathogenesis of ciliopathies.
Article activity feed
-
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
1. General Statements
*We thank the reviewers for their valuable comments. A common suggestion by all reviewers was that the manuscript would benefit from restructuring. Following their recommendation we have restructured this manuscript to improve its readability. *
2. Point-by-point description of the revisions
__Reviewer #1 (Evidence, reproducibility and clarity (Required)): __ The paper from Louka et al. studies the function of Cep104 during the development of Xenopus embryos. They perform overexpression and knock down experiments and address the consequences on neural tube closure, on ciliogenesis, and MT stability and on apical intercalation. …
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
1. General Statements
*We thank the reviewers for their valuable comments. A common suggestion by all reviewers was that the manuscript would benefit from restructuring. Following their recommendation we have restructured this manuscript to improve its readability. *
2. Point-by-point description of the revisions
__Reviewer #1 (Evidence, reproducibility and clarity (Required)): __ The paper from Louka et al. studies the function of Cep104 during the development of Xenopus embryos. They perform overexpression and knock down experiments and address the consequences on neural tube closure, on ciliogenesis, and MT stability and on apical intercalation. There is a lot of data presented on a wide range of topics. While the data on MTs tracks reasonably well with other reports on Cep104, there are some concerns regarding the quality of some of the data and the interpretations based on the experimental results.
Specific Points: It is difficult to assess the effect on apical constriction with the data provided. Please show zoomed in higher mag images. Also this should be coupled with a quantification of cell number and proliferation rates, as it is possible that Cep104 mildly affects proliferation / cell division which could affect cell size. Overall this experiment is not really addressing apical constriction since there is no before and after data. Lots of things could affect apical surface area, most notably proliferation rates which one might predict would be affected by subtle changes to MT dynamics.
__Response: __Following the reviewer's recommendation we now show zoomed in higher magnification images to more clearly demonstrate the larger cell surface area in the morpholino injected neural plate compared to the control non-injected side in the same embryo. We agree with the reviewer that defects in cell proliferation could affect the cell size. If the effect of Cep104 on the cell surface area is caused by defects in cell proliferation, then we would expect this phenotype to persist in other tissues such as the ectoderm. However, we show that this phenotype is specific to the neural plate. On the other hand, if the cell surface area defect is caused by defects in apical constriction, we would expect this phenotype to be stage specific. Following the reviewer's recommendation, we compared the surface area of neuroectoderm cells before and after extensive apical constriction takes. The new data is shown in Figure S2. Our results show no difference in the surface area of neuroectoderm cells in control tracer injected and morpholino injected neuroepithelial cells at stage 13, before extensive apical constriction whereas significant differences are observed in stage 15 embryos during which cells undergo apical constriction. This data strengthens our conclusion that downregulation of Cep104 affects apical constriction.
"This defect was rescued with expression of exogenous human CEP104-GFP mRNA (300pg mRNA) (Figure 1D-E)." This was partially rescued as the control and the rescue are significantly different.
__Response: __We thank the reviewer for this important clarification. We edited the text to more clearly reflect our data.
I am unclear what is being depicted in Figure 1F and G. What is the intense red staining? Is that the blastopore? Which would imply that the stage of analysis is quite different between C and F which is concerning. The same stages should be used.
__Response: __This is an image of the anterior most region of a stage 15 embryo. Occasionally some embryos do display intense phalloidin staining at the neural plate. We replaced the image with a more clear one and moved this data to Figure S2C.
S1A has a boxed region as if there was going to be a zoomed in image, but there is not. It would be nice to see it zoomed in. While the localization is indeed at the base and tips of cilia the base looks too dispersed and big to be the basal body?
__Response: __Following the reviewer's recommendation we now show a zoomed in image of a primary cilium. The boxed area in figure S2A shows the cilium that was used to generate the fluorescence intensity profile plot shown in S2B. The Cep104 signal at the basal body is much stronger compared to the ciliary tip signal. Exposure that allows simultaneous detection of both the base and the tip signal results in overexposure of the signal at the base. This is consistent with observations in primary cilia in cell culture (please refer to Figure 4 in Frikstad et al. 2019 and Figure 3 in Yamazoe et al 2020).
In other systems the depletion of Cep104 decreases primary cilia length. While the authors claim that neural tube cilia are normal there is no quantification to support that and the provided image is hard to assess.
__Response: __Following the reviewer's recommendation we now show quantifications of the length of floor plate cilia (Figure S3C). Floor plate cilia are longer than the cilia found elsewhere in the neural tube. This inherent variability in the length of cilia will likely prevent the detection of small changes in the cilium length elicited by downregulation of Cep104. Therefore, we chose to examine the length of floor plate cilia only, in control and morpholino injected cells. Our results show that downregulation of Cep104 leads to the formation of shorter floor plate cilia which is in agreement with published data in other systems.
While the authors claim broad expression in humans and MO effects in cells without cilia, there is little data supporting the expression of Cep104 in the Xenopus cells being assayed (e.g. goblet cells).
__Response: __We agree with the reviewer that there is little evidence supporting the expression of Cep104 in Xenopus goblet cells. Cep104 is a very low abundance protein and thus very difficult to detect it at endogenous levels For example, Ryniawec et al. (2023) raised an antibody against Drosophila Cep104 that failed to detect the native (endogenous) protein via western blot or immunofluorescence, but successfully recognized the overexpressed (transgenic) Cep104. A proteomic study by Peshkin et al. 2019 showed that Cep104 levels remain relatively constant throughout Xenopus development suggesting that this protein is expressed ubiquitously. This data is shown in Figure 4 where we plot the relative expression levels of Cep104 along with two motile cilia specific genes: hydin and RSPH9.
The data in Figure 2 regarding the explants is difficult to understand and I think missing some key data. The text refers to the level of Gli increasing in the BF injected explants compared to uninjected explants, but the presentation of that is odd as the levels are normalized against uninjected rather than directly compared. And there are no stats for this key experiment. However, I think a bigger concern is the lack of information regarding the presence of cilia. While elongation and Sox2 expression are important they don't address if this tissue is similar to the neural tube in terms of cilia which is key to the interpretations.
__Response: __Following the reviewer's recommendation we changed the presentation of this data. GLI1 levels are now normalized to XBF2 injected explants. The results are the same, Gli1 levels are 25% lower in morphant XBF2 explants (ttest pWe understand the reviewer's concern regarding the presence of cilia in the explants. To our knowledge there are currently no reports on the presence of cilia in the neural ectoderm in Xenopus. We have made several attempts to determine if cilia are present in this tissue during neurulation. However, we have not been able to detect cilia based on immunofluorescence staining for acetylated tubulin and Arl13b in the neural ectoderm. We conclude from this experiment that downregulation of Cep104 negatively affects hedgehog signaling and it remains to be addressed whether this is due to defects in primary cilia.
The localization of Cep104 GFP in the epidermis and the neuroepithelium does not look similar as stated. Ones does not really see the punctate pattern in the neuroectoderm.
Response: We thank the reviewer for pointing this out. To more clearly present this data we now show a plot of the fluorescence profile of Cep104-GFP along cell-cell junctions to demonstrate the punctate localization in the neuroepithelium.
The experiments linking Cep104 to the tips of paused MTs is not particularly convincing. The depolymerization of MTs with nocodazole, will decrease all MTs as well as MT trafficking which could affect Cep104. Comparing this experiment with taxol treatment to stabilize MTs (and decrease dynamics) would be more convincing. Plus the image provided does not support the claim that the leftover EMTB is marked with Cep104.
__Response: __Following the reviewer's recommendation we have examined the effect of taxol on the density of Cep104 apical puncta. We injected embryos with CEP104-GFP and EMTB-scarlet and exposed them to 20 μm taxol and imaged them live at stage 38. Embryos non treated with taxol served as the control. As shown in Figure S4 treatment with taxol led to an increase in the density of Cep104 puncta. This further supports our conclusion that Cep104 localizes to the ends of stable or paused microtubules. We also revised Figure 5 to more clearly show that Cep104 remains associated with the ends of nocodazole resistant EMTB labeled microtubules.
The data in Figure 6 is very difficult to interpret / believe. The quantified effects on MTs are pretty subtle (which is fine...that is why you quantify), but the massive experimental variability questions the meaningfulness of those quantifications. In Fig 6B There are cells with lots of MTs right next to cells with no MTs and both have similar expression levels of Cep104. The staining just doesn't look consistent enough to accurately quantify. Also the effect of Nocodozole on MT stability is quite rapid, on the order of seconds to minutes, it is unclear what ON treatment with nocodazole would even be measuring since in that time there would be lots of secondary effects.
__Response: __We thank the reviewer for this comment. Some cells in the epidermis lack apical microtubules as the reviewer correctly points out. Cells without strong apical microtubule staining are seen in both control and morpholino injected cells. Here we quantified the number of control and morphant cells per embryo that lack apical microtubules (DMSO treated embryos). Our results show that similar numbers of control and morphant cells per embryo appear to lack apical microtubules. We think that the heterogeneity in tubulin signal is not an artifact of immunofluorescence staining since these cells are adjacent to cells with clear tubulin staining. Although the source of this variability is still unknown, the fact that an equal number of control and morphant cells show this phenotype suggests that this is unlikely to be linked to the injections or drug treatment. Those cells were excluded from the quantifications shown in Figures 6C and 6D It is possible that these cells are preparing to enter mitosis.
We think that the reviewer refers to the acute effects of nocodazole seen in cell cultures. However, in Xenopus tadpoles we didn't observe any effect on microtubules after short nocodazole treatment at low temperatures.
The authors propose that overexpressing Cep104 would lead to stabilized MTs which is a reasonable hypothesis, however, they test this in multiciliated cells that already have a ton of acetylated MTs. If their hypothesis is correct it should lead to an increase in acetylated tubulin in non multiciliated cells which don't have much to begin with. This would be a marked improvement as the side projection quantification seems a little suspect as the analysis requires a precises ROI that eliminates the strong cilia acetylation staining. While I believe that could be done, the image provided looks as if it might cut off some of the apical surface which highlights the challenge.
__Response: __Following the reviewer's recommendation, we examined the effect of Cep104 overexpression in non-MCCs on Xenopus epidermis. We show in Figure 7 that overexpression of Cep104 leads to a significant increase in the levels of acetylated tubulin in the cytoplasm of non-MCCs. We also show that overexpression of GFP alone did not have an effect on microtubule acetylation (Figure S5A). We moved the data on the cytoplasmic levels of acetylated microtubules in MCCs to figure S5B. We would like to clarify that the ROI to mark the cell body of MCCs was drawn right below the apical phalloidin signal to ensure that no signal derived from motile cilia will be included in the quantifications. A more detailed explanation of the quantification methods is included in this revised manuscript.
Minor: Overall the color choice of images does not conform to the color blind favorable options that are becoming standard in the field. Also to the extent possible the colors should be consistent (e.g. Fig 4 A Cep104-GFP is green but in B it is red).
__Response: __We thank the reviewer for this comment. We have changed the color choices in the figures to conform to the color blind.
The recent Xenopus Cep104 paper was referenced with two references, and the wording of those two sentences was redundant.
__Response: __We thank the reviewer for this comment. We edited the text accordingly.
__Reviewer #2 (Evidence, reproducibility and clarity (Required)): __ This study by Louka et al., investigates the function of Cep104, a protein associated with Joubert syndrome, in Xenopus. Several aspects are studied at different scales. Loss of function of this protein suggests a role in neural tube closure, apical constriction, and HH signaling. Moving on in the study, the authors investigate the localization of Cep104 in the primary cilia of the neural tube before focusing on its localization in multiciliated cells. They then look at the consequences of loss of function on motile cilia and conclude that it plays a role in the length of the distal segment. They then show an association of Cep104 with cytoplasmic microtubules in non-multiciliated cells of the Xenopus epidermis. They then analyze the function of Cep104 on these microtubules and show that loss of Cep104 function increases the speed of EB1 comets. They then looked at the impact of loss of function on microtubule stability and finally the impact of gain of function. Finally, they returned to the multiciliated cells and described an intercalation defect that correlated with decreases in acetylated tubulin. I think that certain controls are missing and that the choice of illustrations should be reconsidered (better quality, appropriate zoom). In terms of form, the text is not easy to read and the manuscript would benefit from reformatting to highlight the logical links between the different experiences and avoid a catalog-like effect. I would advise the authors to revise their introduction to make it less disjointed and guide readers toward the questions addressed by the manuscript.
Response: We thank the reviewer for the constructive criticism. We have revised the introduction to make it easier to read.
Below are specific comments and remarks: Figure 1: Why the conclusion is a "delay" in neural tube closure? At what stage is this analyzed? Is there a recovery of NT closure at later stage? A: I would suggest to provide control picture of non-injected and tracer only injected embryos. B: Statistics are missing on the graph D: mention what was injected instead of "+ rescue". Close up picture would allow a better appreciation of the differences in surface area.
Response: We thank the reviewer for this comment. The image shown in Figure 1A is from late neurula embryos, stage 18. We conclude that it is a delay in neural tube closure because the neural tube does close and the embryos develop to tailbud stages. To demonstrate the delay in neural tube closure we now include a time lapse sequence of a neurula stage embryo injected with the morpholino unilaterally which shows that the morpholino injected side moves towards the midline slower compared to the control uninjected side (movie 1). We also included a representative image of the dorsal side of a tailbud embryo injected unilaterally with the CEP104 morpholino to show that the neural tube has closed and the embryos develop to tailbud stages (figure S1D).
Following the reviewer's recommendation, we also show images of embryos injected unilaterally with the tracer alone (Figure S2), we included the statistical analysis for graph 1D, revised image 1D to show that the embryo is injected with the morpholino and CEP104-GFP and provide close ups to allow for better appreciation of the differences in surface area.
Figure S1: To illustrate the claim that cilia are not affected, it would be good to show injection of tracer alone and compare to tracer + morpholino. Also, to provide a measure of the cilia size.
__Response: __Following the reviewer's recommendation we quantified the length of floor plate cilia in the neural tube of control and morpholino injected embryos. As explained in our response to a comment by reviewer 1, the floor plate cilia are longer than the cilia found elsewhere in the neural tube. This inherent variability in the length of cilia will likely prevent the detection of small changes in the cilium length elicited by downregulation of Cep104. Therefore, we chose to examine the length of floor plate cilia only in control and morpholino injected cells. Our results show that downregulation of Cep104 leads to the formation of shorter floor plate cilia which is in agreement with published data in other systems (Figure S3C).
Figure 2: Please provide pictures to illustrate graph D.
__Response: __The graph in Figure 2D shows RT-qPCR results for CEP104 in BF2 and BF2 and morpholino injected explants as compared to non-injected explants. We do not have a working antibody that would allow us to show the downregulation at the protein level.
Figure 5: "Interestingly, most of the nocodazole-resistant stable microtubules were positive for Cep104 (Figure 5C, arrows). " The variation in density of Cep104-GFP signal is not visible on the pictures provided in C. I would suggest to show higher magnifications. Also, in the DMSO treated picture the Cep104GFP signal looks really different when compared to Cep104-GFP signal shown in B. Arrows should be reported on all channels. However, it not clear what we should see with this arrows. 5C: it seems that in nocodazole treated condition the Cep104-GFP is at the cilia base in MCCs which is different from the DMSO control condition. The basal body signal was not seen in the figure 3A which analyze the localization of Cep104-GFP in MCCs. Why not comment on this? Is it a phenotype on MCCs ?
__Response: Following the reviewer's recommendation, we now show higher magnifications of the images shown in Figure 5C. We removed the arrows as most reviewers found them confusing. To demonstrate the presence of Cep104 at the ends of nocodazole resistant EMTB labeled microtubules we show zoomed images and a representative fluorescence intensity profile plot. __Figure 5B shows an image of a non-MCC whereas Figure 5C shows a larger area on the tadpole epidermis which includes both MCCs and non-MCCs. We thank the reviewer for pointing out that the localization of Cep104 in 5C looks different from 3A. We do not think this is a phenotype on MCCs. In Figure 3A we imaged only the tips of cilia which is why it looks different from 5C in which we imaged the apical surface of the cells as well. We disagree with the reviewer regarding the comment *'5C: it seems that in nocodazole treated condition the Cep104-GFP is at the cilia base in MCCs which is different from the DMSO control condition'. *The basal body localization of Cep104 is shown in the DMSO image as well. We hope that it will be clear in this revised figure.
Figure 6: Intriguingly, morphant non-MCCs have significantly more mean β-tubulin signal compared to control non-MCCs in embryos treated with DMSO (Figure 6C). impossible to appreciate on the figures. Please specify on the figure what is considered as a morphant non-MCC versus a control non-MCC. The membrane-cherry positive cells (supposedly morphant? it has to be clarified show very heterogenous tubulin expression) If the point here is to show that microtubules are more sensitive to nocodazole in morphant cells as compared to control. I would suggest to show all conditions on a same graph. At least annotate more the graph for a self-explanatory figure (DMSO , Nocodazole).
__Response: __We agree with the reviewer that it impossible to appreciate the difference in β-tubulin signal between control and morphant non-MCCs. Based on the quantifications of mean β-tubulin fluorescence intensity there is 5% difference in the fluorescence intensity between the two groups. Statistical analysis using t-test shows that although very small, this difference is statistically significant which is why we mention it in the manuscript. We have removed this statement and data from the revised manuscript because this is a very subtle phenotype, and it is beyond the scope of this experiment.
Following the reviewer's recommendation, we clarify that mem-cherry positive cells contain the morpholino and mem-cherry negative cells are the control cells. We marked with a white asterisk the morphant non-MCCs. To address the heterogenous tubulin levels we provide quantifications which show that a similar number of control and morphant cells appear to lack microtubules. We think that the heterogeneity in tubulin signal is not an artifact of immunofluorescence staining since these cells are adjacent to cells with clear tubulin staining. Although the source of this variability is still unknown, the fact that an equal number of control and morphant cells show this phenotype suggests that this is unlikely to be linked to the injections or drug treatment. Those cells were excluded from the quantifications shown in figure 6. It is possible that these cells are preparing to enter mitosis. The reviewer is correct; the point of this experiment is to examine the effect of Cep104 downregulation on the sensitivity of microtubules to nocodazole. To more clearly present the results of this experiment we normalize the β-tubulin fluorescence Intensity in morphant cells to the one in control cells in the same embryo and we compare the normalized intensity in DMSO and nocodazole treated embryos.
Figure 7: Statistics are missing on Graph B
__ ____Response: __Following the reviewer's recommendation, we added the statistics on the graph.
Comment on the text: "Cep104 signal shows the characteristic two dot pattern in motile cilia (Figure 3A) that was also observed in a recent study using Xenopus Cep10465 and in the cilia of Tetrahymena50. This is in agreement with a recent study showing the characteristic two dot pattern for Xenopus Cep104 as well66 " ref 65 and 66b are the same (Hong et al., preprint)
__ ____Response: __We thank the reviewer for pointing this out. We edited the text to avoid repetition and corrected the references.
"This data suggests that downregulation of CEP104 affects the stability of cytoplasmic microtubules." I would suggest a more precise conclusion by stating how is it affected? More stable? Less stable? Important for the follow-up demonstration.
__ ____Response: We edited the text according to the reviewer's recommendation to precisely conclude that downregulation of Cep104 makes cytoplasmic microtubules less stable. __
Movies: Please annotate properly movie 2 and 3 so the reader can know what he/she is looking.
__Response: __Following the reviewer's comment, we revised the movie annotations to help the reader know what they are looking.
__Reviewer #3 (Evidence, reproducibility and clarity (Required)): __ The manuscript entitled "Ciliary and non-ciliary functions of Cep104 in Xenopus" by Louka et al. investigate roles for the centriole and cilia tip protein Cep104 in Xenopus embryos. The authors show that depletion of Cep104 prevents neural tube closure due to inefficient apical constriction of neural cells and defective hedgehog signaling. Cep104 depletion also resulted in structural and functional ciliary defects in multi-ciliated cells. Surprisingly, the authors discover a role for Cep104 in stabilizing cytoplasmic microtubules in non-ciliated and multi-ciliated cells. Reduced microtubule stability in Cep104-depleted cells correlated with reduced apical intercalation of multi-ciliated cells in the epidermis.
Overall, I find this manuscript difficult to understand because the experiments lack description of the findings within a normal developmental context and the findings are not developed into a cohesive narrative. I do find the study to be potentially impactful as the authors characterize Cep104 in a novel system (previous peer-reviewed studies have investigated Cep104 in human cell lines, Drosophila, zebrafish, Tetrahymena, and Chlamydomonas) with disease-relevant biology (neural development); however, mechanistic links are not properly explored. Over the course of their investigation, the authors made the novel finding that Cep104 controls the dynamics of cytoplasmic microtubules. However, this is not directly tested and potential pleiotropic effects of the developmental defects caused by Cep104 depletion confound the results.
Response: We thank the reviewer for their comments. We tried to address this by restructuring the manuscript to describe the results in more detail within a normal developmental context.
Major Critiques: The developmental context of experiments is not made clear. The authors use different tissues at varying developmental stages to perform experiments. However, these findings are not explored in depth and, therefore, the manuscript does not advance our understanding of Cep104's role in any of the processes explored.
__ ____Response:__ We thank the reviewer for their comment. We took advantage of different tissues during Xenopus development to understand the cellular and molecular function of this protein in vivo. In this manuscript we show that Cep104 is involved in neural tube closure likely through its effect on apical constriction. Our data show that Cep104 is important for the stability of cytoplasmic microtubules and this is further demonstrated through its role in apical intercalation of multiciliated cells, a process known to depend on stable microtubules. Although our data do not advance our understanding on developmental processes such as apical constriction and MCC apical intercalation, they do improve our understanding of how Cep104 impacts cytoplasmic microtubules which has not been addressed in vivo yet.
While the potential role of Cep104 in cytoplasmic microtubule regulation is intriguing, the experiments in the manuscript do not directly test this function. Because Cep104 depletion appears to have a profound developmental effect, it is difficult to interpret changes to EB1 velocity as directly attributed to Cep104 function. Additionally, the only evidence for Cep104 localization occurs in cells overexpressing human Cep104. The authors must directly visualize endogenous Cep104 to conclude microtubule or membrane localization, which they can also use to demonstrate Cep104 depletion in the morpholino experiments. Additionally, the assertion that Cep104 is binding plus-ends of cytoplasmic microtubules is not experimentally supported.
__ ____Response: __Unfortunately, we cannot directly visualize endogenous Cep104 because there is no commercially available antibody that works in Xenopus. Cep104 is a very low abundance protein, and this is highlighted in the study by John M.Ryniawec et al. 2023, where they generated an antibody against the drosophila Cep104 which detected the GFP-tagged DmCep104 but failed to detect the endogenous protein. Given that the ciliary and basal body signal of Cep104 represents the cumulative signal from nine microtubules, one can appreciate the difficulty of observing the Cep104 signal in individual microtubules. None of the commercially available Cep104 antibodies that we have tested worked against the Xenopus protein in immunofluorescence or western blot experiments. We agree with the reviewer that we do not experimentally test the binding of Cep104 to the microtubule plus-end. This has been demonstrated by others. In Jiang et al. 2012 it was showed that GFP-Cep104 co-immunoprecipitates with GST-EB1 but not with GST-EB1 that lacks the tail which contains the SxIP binging motif. In Yamazoe et al. 2020 study it was shown that exogenous Cep104 co-immunoprecipitates with exogenous EB1 and Cep104 with mutated SxIP motif (SKNN) fails to co-immunoprecipitate with EB1. This shows that Cep104 interacts with EB1 through its SxIP motif. In addition, overexpression of Cep104 recruits Cep97 to microtubule tips suggesting that it acts as a +TIP protein. A recent study by Saunders et al. 2025 showed that in in vitro microtubule reconstitution assays, Cep104 could not autonomously bind the microtubule plus-end at low concentrations but in the presence of EB3 it could bind the microtubule plus-end and block microtubule polymerization at the same low concentration. This shows that Cep104 interacts with EB3, localizes to the microtubule plus-end and affects its dynamics in vitro. We added this information in the manuscript to more clearly show that the interaction of Cep104 and EB proteins is well documented. We anticipate that this interaction will hold true in all cell types where the two proteins are co-expressed.
Additional Critiques: Figure S1. I only see the emergence of a shorter product after Cep104 depletion. Should PCR using Exon5-7 still work in successful knockdown? If not, then it is unclear what was quantified to determine Cep104 depletion as morpholino bands appear no different than control.
__ ____Response: __We thank the reviewer for this comment. PCR using exon5-7 will not work when splice blocking by the morpholino takes place. This is a knockdown approach and the efficiency of the morpholino is about 90%. Upon completion of the RT-qPCR cycle the samples were analyzed by gel electrophoresis to demonstrate that 1) alternative splicing took place (see two products with exon 3-7 primers) and 2) the presence of a single product for all primer sets used.
Figure 1A. Is this an example of an open or closed NTC? Show data used to determine the statement "no difference during convergent extension".
__ ____Response: __This is an example of an embryo that was unilatterally injected with the morpholino. The left side is the control non-injected side and the right side is the morpholino injected. We added this information on the figure to make it more self-explanatory. In Figure 2 the elongation of the BF2 injected explants is due to convergent extension. The statement "no difference during convergent extension" was removed from the revised manuscript.
Figure S2C. What does "Does not effect formation of cilia" mean? Does Cep104 depletion does not effect number, length, etc? Show quantitation used to determine this?
__ ____Response:__ Following the reviewer's recommendation, we quantified the length of floor plate cilia in control and morpholino injected embryos. As mentioned in our response to reviewer 1 and 2, floor plate cilia are longer than the cilia found elsewhere in the neural tube. This inherent variability in the length of cilia will likely prevent the detection of small changes in the cilium length elicited by downregulation of Cep104. Therefore, we chose to examine the length of floor plate cilia only, in control and morpholino injected cells. Our results show that downregulation of Cep104 leads to the formation of shorter floor plate cilia which is in agreement with published data in other systems.
Figure 5B. Along with strong Cep104 localization to membranes, there also appears to be strong EMTB localization. Is this also present in beta-tubulin immunostaining? Are these localizing to a cortical population of microtubules or to the membrane?
__ ____Response: __We thank the reviewer for their comment. The Cep104 puncta at the cell periphery, are reduced/lost upon nocodazole treatment thus we conclude that Cep104 localizes to microtubules and not the cell membrane (Figure 5C, zoomed images). Of course, we cannot exclude the possibility that microtubules are required to target CEP104 to the plasma membrane. We edited the text to clearly state this conclusion.
Figure 6C and 6D. These two panels have the same labels. The authors should denote that 6D is in nocodazole-treated explants.
__ ____Response:__ We thank the reviewer for this comment. We edited this figure to more clearly present the results of this experiment: We normalized the β -tubulin levels in morphant cells to that of control cells in the same embryo (mosaic morphant embryos were used in this experiment). The graph shows the mean normalized β -tubulin levels per embryo treated with DMSO or nocodazole.
Figure 7. What are Cep104 levels at stage 18-19?
__ ____Response: __Following the reviewer's comment we now show the Cep104 protein expression levels during Xenopus development as reported on Xenbase (Figure 4). Cep104 is expressed at low levels from gastrulation to tailbud stages (Figure 4D).
-
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
The manuscript entitled "Ciliary and non-ciliary functions of Cep104 in Xenopus" by Louka et al. investigate roles for the centriole and cilia tip protein Cep104 in Xenopus embryos. The authors show that depletion of Cep104 prevents neural tube closure due to inefficient apical constriction of neural cells and defective hedgehog signaling. Cep104 depletion also resulted in structural and functional ciliary defects in multi-ciliated cells. Surprisingly, the authors discover a role for Cep104 in stabilizing cytoplasmic microtubules in non-ciliated and multi-ciliated cells. Reduced microtubule stability in Cep104-depleted cells …
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
The manuscript entitled "Ciliary and non-ciliary functions of Cep104 in Xenopus" by Louka et al. investigate roles for the centriole and cilia tip protein Cep104 in Xenopus embryos. The authors show that depletion of Cep104 prevents neural tube closure due to inefficient apical constriction of neural cells and defective hedgehog signaling. Cep104 depletion also resulted in structural and functional ciliary defects in multi-ciliated cells. Surprisingly, the authors discover a role for Cep104 in stabilizing cytoplasmic microtubules in non-ciliated and multi-ciliated cells. Reduced microtubule stability in Cep104-depleted cells correlated with reduced apical intercalation of multi-ciliated cells in the epidermis.
Overall, I find this manuscript difficult to understand because the experiments lack description of the findings within a normal developmental context and the findings are not developed into a cohesive narrative. I do find the study to be potentially impactful as the authors characterize Cep104 in a novel system (previous peer-reviewed studies have investigated Cep104 in human cell lines, Drosophila, zebrafish, Tetrahymena, and Chlamydomonas) with disease-relevant biology (neural development); however, mechanistic links are not properly explored. Over the course of their investigation, the authors made the novel finding that Cep104 controls the dynamics of cytoplasmic microtubules. However, this is not directly tested and potential pleiotropic effects of the developmental defects caused by Cep104 depletion confound the results.
Major Critiques:
The developmental context of experiments is not made clear. The authors use different tissues at varying developmental stages to perform experiments. However, these findings are not explored in depth and, therefore, the manuscript does not advance our understanding of Cep104's role in any of the processes explored.
While the potential role of Cep104 in cytoplasmic microtubule regulation is intriguing, the experiments in the manuscript do not directly test this function. Because Cep104 depletion appears to have a profound developmental effect, it is difficult to interpret changes to EB1 velocity as directly attributed to Cep104 function. Additionally, the only evidence for Cep104 localization occurs in cells overexpressing human Cep104. The authors must directly visualize endogenous Cep104 to conclude microtubule or membrane localization, which they can also use to demonstrate Cep104 depletion in the morpholino experiments. Additionally, the assertion that Cep104 is binding plus-ends of cytoplasmic microtubules is not experimentally supported.
Additional Critiques:
Figure S1. I only see the emergence of a shorter product after Cep104 depletion. Should PCR using Exon5-7 still work in successful knockdown? If not, then it is unclear what was quantified to determine Cep104 depletion as morpholino bands appear no different than control.
Figure 1A. Is this an example of an open or closed NTC? Show data used to determine the statement "no difference during convergent extension".
Figure S2C. What does "Does not effect formation of cilia" mean? Does Cep104 depletion does not effect number, length, etc? Show quantitation used to determine this?
Figure 5B. Along with strong Cep104 localization to membranes, there also appears to be strong EMTB localization. Is this also present in beta-tubulin immunostaining? Are these localizing to a cortical population of microtubules or to the membrane?
Figure 6C and 6D. These two panels have the same labels. The authors should denote that 6D is in nocodazole-treated explants.
Figure 7. What are Cep104 levels at stage 18-19?
Significance
Overall, I find this manuscript difficult to understand because the experiments lack description of the findings within a normal developmental context and the findings are not developed into a cohesive narrative. I do find the study to be potentially impactful as the authors characterize Cep104 in a novel system (previous peer-reviewed studies have investigated Cep104 in human cell lines, Drosophila, zebrafish, Tetrahymena, and Chlamydomonas) with disease-relevant biology (neural development); however, mechanistic links are not properly explored. Over the course of their investigation, the authors made the novel finding that Cep104 controls the dynamics of cytoplasmic microtubules. However, this is not directly tested and potential pleiotropic effects of the developmental defects caused by Cep104 depletion confound the results.
-
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 Louka et al., investigates the function of Cep104, a protein associated with Joubert syndrome, in Xenopus. Several aspects are studied at different scales. Loss of function of this protein suggests a role in neural tube closure, apical constriction, and HH signaling. Moving on in the study, the authors investigate the localization of Cep104 in the primary cilia of the neural tube before focusing on its localization in multiciliated cells. They then look at the consequences of loss of function on motile cilia and conclude that it plays a role in the length of the distal segment. They then show an association of Cep104 …
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 Louka et al., investigates the function of Cep104, a protein associated with Joubert syndrome, in Xenopus. Several aspects are studied at different scales. Loss of function of this protein suggests a role in neural tube closure, apical constriction, and HH signaling. Moving on in the study, the authors investigate the localization of Cep104 in the primary cilia of the neural tube before focusing on its localization in multiciliated cells. They then look at the consequences of loss of function on motile cilia and conclude that it plays a role in the length of the distal segment. They then show an association of Cep104 with cytoplasmic microtubules in non-multiciliated cells of the Xenopus epidermis. They then analyze the function of Cep104 on these microtubules and show that loss of Cep104 function increases the speed of EB1 comets. They then looked at the impact of loss of function on microtubule stability and finally the impact of gain of function. Finally, they returned to the multiciliated cells and described an intercalation defect that correlated with decreases in acetylated tubulin. I think that certain controls are missing and that the choice of illustrations should be reconsidered (better quality, appropriate zoom). In terms of form, the text is not easy to read and the manuscript would benefit from reformatting to highlight the logical links between the different experiences and avoid a catalog-like effect. I would advise the authors to revise their introduction to make it less disjointed and guide readers toward the questions addressed by the manuscript.
Below are specific comments and remarks:
Figure 1:
Why the conclusion is a "delay" in neural tube closure? At what stage is this analyzed? Is there a recovery of NT closure at later stage? A: I would suggest to provide control picture of non-injected and tracer only injected embryos. B: Statistics are missing on the graph D: mention what was injected instead of "+ rescue". Close up picture would allow a better appreciation of the differences in surface area.
Figure S1:
To illustrate the claim that cilia are not affected, it would be good to show injection of tracer alone and compare to tracer + morpholino. Also, to provide a measure of the cilia size.
Figure 2:
Please provide pictures to illustrate graph D.
Figure 5:
"Interestingly, most of the nocodazole-resistant stable microtubules were positive for Cep104 (Figure 5C, arrows). "
- The variation in density of Cep104-GFP signal is not visible on the pictures provided in C. I would suggest to show higher magnifications. Also, in the DMSO treated picture the Cep104GFP signal looks really different when compared to Cep104-GFP signal shown in B. Arrows should be reported on all channels. However, it not clear what we should see with this arrows. 5C: it seems that in nocodazole treated condition the Cep104-GFP is at the cilia base in MCCs which is different from the DMSO control condition. The basal body signal was not seen in the figure 3A which analyze the localization of Cep104-GFP in MCCs. Why not comment on this? Is it a phenotype on MCCs ? Figure 6: Intriguingly, morphant non-MCCs have significantly more mean β-tubulin signal compared to control non-MCCs in embryos treated with DMSO (Figure 6C).
- impossible to appreciate on the figures. Please specify on the figure what is considered as a morphant non-MCC versus a control non-MCC. The membrane-cherry positive cells (supposedly morphant? it has to be clarified show very heterogenous tubulin expression)
If the point here is to show that microtubules are more sensitive to nocodazole in morphant cells as compared to control. I would suggest to show all conditions on a same graph. At least annotate more the grap for a self-explanatory figure (DMSO , Nocodazole). Figure 7: Statistics are missing on Graph B Comment on the text: "Cep104 signal shows the characteristic two dot pattern in motile cilia (Figure 3A) that was also observed in a recent study using Xenopus Cep10465 and in the cilia of Tetrahymena50. This is in agreement with a recent study showing the characteristic two dot pattern for Xenopus Cep104 as well66 "
- ref 65 and 66b are the same (Hong et al., preprint)
"This data suggests that downregulation of CEP104 affects the stability of cytoplasmic microtubules."
- I would suggest a more precise conclusion by stating how is it affected? More stable? Less stable? Important for the follow-up demonstration.
Movies:
Please annotate properly movie 2 and 3 so the reader can know what he/she is looking.
Referees cross-commenting
Similar feeling that reviews are consistent
Significance
This study investigates the role of the proprotein Cep104 in Xenopus. Cep104 is a protein associated with Joubert syndrome, whose role in primary cilia has been extensively documented. While its localization at the tip of motile cilia has also been reported, this study provides functional evidence for the role of Cep104 in motile cilia. In addition, the study looks at the role of Cep104 on non-cilial microtubules, which is the original aspect of the paper and may ultimately lead to a better understanding of Joubert syndrome. However, I believe that the evidence provided (controls, illustrations) needs to be improved. This paper will be of interest to a specialized audience with an interest in proteins associated with cilia and microtubules.
I am a cell biologist specialized in the study of multiciliated cells using advanced imaging methods and Xenopus and mice as models. I believe my expertise was a perfect match for this manuscript.
-
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 paper from Louka et al. studies the function of Cep104 during the development of Xenopus embryos. They perform overexpression and knock down experiments and address the consequences on neural tube closure, on ciliogenesis, and MT stability and on apical intercalation. There is a lot of data presented on a wide range of topics. While the data on MTs tracks reasonably well with other reports on Cep104, there are some concerns regarding the quality of some of the data and the interpretations based on the experimental results.
Specific Points:
It is difficult to assess the effect on apical constriction with the data provided. …
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 paper from Louka et al. studies the function of Cep104 during the development of Xenopus embryos. They perform overexpression and knock down experiments and address the consequences on neural tube closure, on ciliogenesis, and MT stability and on apical intercalation. There is a lot of data presented on a wide range of topics. While the data on MTs tracks reasonably well with other reports on Cep104, there are some concerns regarding the quality of some of the data and the interpretations based on the experimental results.
Specific Points:
It is difficult to assess the effect on apical constriction with the data provided. Please show zoomed in higher mag images. Also this should be coupled with a quantification of cell number and proliferation rates, as it is possible that Cep104 mildly affects proliferation / cell division which could affect cell size. Overall this experiment is not really addressing apical constriction since there is no before and after data. Lots of things could affect apical surface area, most notably proliferation rates which one might predict would be affected by subtle changes to MT dynamics.
"This defect was rescued with expression of exogenous human CEP104-GFP mRNA (300pg mRNA) (Figure 1D-E)." This was partially rescued as the control and the rescue are significantly different.
I am unclear what is being depicted in Figure 1F and G. What is the intense red staining? Is that the blastopore? Which would imply that the stage of analysis is quite different between C and F which is concerning. The same stages should be used.
S1A has a boxed region as if there was going to be a zoomed in image, but there is not. It would be nice to see it zoomed in. While the localization is indeed at the base and tips of cilia the base looks too dispersed and big to be the basal body?
In other systems the depletion of Cep104 decreases primary cilia length. While the authors claim that neural tube cilia are normal there is no quantification to support that and the provided image is hard to assess.
While the authors claim broad expression in humans and MO effects in cells without cilia, there is little data supporting the expression of Cep104 in the Xenopus cells being assayed (e.g. goblet cells).
The data in Figure 2 regarding the explants is difficult to understand and I think missing some key data. The text refers to the level of Gli increasing in the BF injected explants compared to uninjected explants, but the presentation of that is odd as the levels are normalized against uninjected rather than directly compared. And there are no stats for this key experiment. However, I think a bigger concern is the lack of information regarding the presence of cilia. While elongation and Sox2 expression are important they don't address if this tissue is similar to the neural tube in terms of cilia which is key to the interpretations.
The localization of Cep104 GFP in the epidermis and the neuroepithelium does not look similar as stated. Ones does not really see the punctate pattern in the neuroectoderm.
The experiments linking Cep104 to the tips of paused MTs is not particularly convincing. The depolymerization of MTs with nocodazole, will decrease all MTs as well as MT trafficking which could affect Cep104. Comparing this experiment with taxol treatment to stabilize MTs (and decrease dynamics) would be more convincing. Plus the image provided does not support the claim that the leftover EMTB is marked with Cep104.
The data in Figure 6 is very difficult to interpret / believe. The quantified effects on MTs are pretty subtle (which is fine...that is why you quantify), but the massive experimental variability questions the meaningfulness of those quantifications. In Fig 6B There are cells with lots of MTs right next to cells with no MTs and both have similar expression levels of Cep104. The staining just doesn't look consistent enough to accurately quantify. Also the effect of Nocodozole on MT stability is quite rapid, on the order of seconds to minutes, it is unclear what ON treatment with nocodazole would even be measuring since in that time there would be lots of secondary effects.
The authors propose that overexpressing Cep104 would lead to stabilized MTs which is a reasonable hypothesis, however, they test this in multiciliated cells that already have a ton of acetylated MTs. If their hypothesis is correct it should lead to an increase in acetylated tubulin in non multiciliated cells which don't have much to begin with. This would be a marked improvement as the side projection quantification seems a little suspect as the analysis requires a precises ROI that eliminates the strong cilia acetylation staining. While I believe that could be done, the image provided looks as if it might cut off some of the apical surface which highlights the challenge.
Minor:
Overall the color choice of images does not conform to the color blind favorable options that are becoming standard in the field. Also to the extent possible the colors should be consistent (e.g. Fig 4 A Cep104-GFP is green but in B it is red).
The recent Xenopus Cep104 paper was referenced with two references, and the wording of those two sentences was redundant.
Referees cross-commenting
I feel that all three reviews are pretty consistent and I do not have any issues with the other reviews.
Significance
Strengths. Cep104 appears to be a hot topic right now as there are several papers in bioRXiv. I suspect that this led to a bit of a rushed submission. The other papers focus mostly on understanding the mechanisms of the ciliary roles of Cep104 which is well established. In other systems the broad phenotypes associated with Cep104 depletion are assumed to be through loss of cilia mediated HH signaling. This paper proposes a number of non ciliary roles for Cep104 which given its broad distribution could be relevant. If true these findings would add considerably to the field. Given that MTs do lots of things other than make cilia it would not be too surprising for Cep104 to have MT specific phenotypes as proposed here.
Weaknesses. The quality of much of the data makes it difficult to assess the claims of broad importance. Key experiments critical to the interpretation of the data are lacking.
-
