Notch-signaling is required for mediating between two pattern forming processes during head regeneration in Hydra polyps

  1. Mona Steichele
  2. Lara Sauermann
  3. Qin Pan
  4. Jasmin Moneer
  5. Alexandra de la Porte
  6. Martin Heß
  7. Moritz Mercker
  8. Catharina Strube
  9. Heinrich Flaswinkel
  10. Marcell Jenewein
  11. Angelika Böttger

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Hydra polyps regenerate lost body parts, including the head. In addition, Hydra head tissue has organizer properties thus being able to recruit body column tissue from a host polyp to produce ectopic hydranths after transplantation. These pattern forming processes involve Notch- and Wnt/β-catenin-signaling. Hydra head regeneration consists of two parts, hypostome/organizer and tentacle development. Previous work had shown that the Notch inhibitor DAPT blocks hypostome regeneration and organizer formation, but not the appearance of tentacle genes and tentacle tissue. Here we show that the β-catenin inhibitor iCRT14 blocks tentacle regeneration, but not regeneration of hypostome and organizer tissue. Using RT-qPCR gene expression analyzes during head regeneration we found that DAPT inhibits HyWnt3 - and HyBMP2/4 expression and expression of transcriptional repressor genes including CnGsc , Sp5 and HyHes, while increasing expression of HyBMP5/8b and the c-fos -related gene HyKayak. ICRT14 blocks expression of the tentacle specification factor HyAlx, but not expression of HyWnt3 . Thus, in accordance with regeneration of two head structures we find two signaling and gene expression modules with HyWnt3 and HyBMP4 part of a hypostome/organizer module, and BMP5/8 , HyAlx and β-catenin part of a tentacle module. We conclude that Notch functions as an inhibitor of tentacle production to allow regeneration of hypostome/head organizer. Furthermore, with HyKayak we present a candidate target gene for HvNotch induced repressor genes. Using siRNA and the Fos/Jun-inhibitor T5224 we show that HyKayak attenuates the expression of HyWnt3. Finally, Notch signaling was not required for head regeneration of fresh water polyps of Craspedacusta . Polyps of Craspedacusta do not have tentacles and thus, after head removal only regenerate a hypostome with a crescent of nematocytes around the mouth opening. This corroborates the idea that Notch-signaling mediates between two pattern forming processes during Hydra head regeneration.

Article activity feed

  1. Version published to 10.1101/2024.02.02.578611v2 on bioRxiv
  2. 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-2024-02378

    Corresponding author(s): Angelika Böttger

    [Please use this template only if the submitted manuscript should be considered by the affiliate journal as a full revision in response to the points raised by the reviewers.

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

    1. General Statements [optional]

    After we have carefully studied the four reviews we have received, we made some major revisions to the manuscript. These included the following main points:

    • Concerns regarding clarity of the manuscript: we have substantially edited the abstract, introduction and discussion part of the manuscript and added many more references to previous work by other authors, especially Cazet 2021, Tursch 2022 and Gahan 2017. We focused our introduction and discussion on organizer function and on the Gierer-Meinhardt-Model for pattern formation. We think that the conclusions are of great general interest because they suggest a function of the *Hydra *head organizer according to the original definition by Hans Spemann, that is “harmonious interlocking of separate processes which makes up development”. Notch signaling, in our interpretation, is an instrument for this function of the organizer. Comparison with *Craspedacusta *compellingly illustrates this idea.
    • Concerns regarding Craspedacusta experiments: we have isolated four Craspedacusta transcripts (CsSp5, CsWnt3, CsAlx and CsNOWA) and analyzed their response to DAPT during head regeneration in Craspedacusta. This revealed that DAPT did not inhibit *CsWnt3 *expression, in accordance with it not having an effect on head regeneration in Craspedacusta However, DAPT inhibited expression of the other potential CsNotch target genes, confirming that DAPT generally works in *Craspedacusta *polyps as Notch-inhibitor.
    • Concerns regarding HyKayak function: we have conducted a rescue experiment to show the function of Hykayak as a target for Notch-regulated repressor genes and a local inhibitor of Wnt-3 expression, which revealed that the expected up-regulation of HyWnt3 in DAPT-treated animals was very weak and did not rescue the DAPT-regeneration phenotype-this was discussed, but data were not included.

    2. Point-by-point description of the revisions

    This section is mandatory. *Please insert a point-by-point reply describing the revisions that were already carried out and included in the transferred manuscript. *

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

    Major:* • The introduction is lacking a full description of what is known about transcriptional changes during Hydra regeneration and in particular the role of Wnt signalling in this process. Of note the authors do not cite several important studies from recent years including (but not limited to):*

    *https://doi.org/10.1073/pnas.2204122119 *

    *https://doi.org/10.1186/s13072-020-00364-6 *

    *https://doi.org/10.1101/587147 *

    *https://doi.org/10.7554/eLife.60562 This problem is further compounded later when the authors do not cite/discuss work which has performed the same or similar analyses to their own. The authors should endeavor to give a more comprehensive background knowledge. *

    Answer:

    Our work focuses on the role of Notch-signalling during Hydra head regeneration, specifically when the head is removed at an apical position. We therefore now have included additional information about transcriptional changes during this process in the introduction. In addition, we have included the suggested citations in the introduction to give a more general background knowledge.

    e.g. .Following decapitation, the expression of Hyβ-catenin and HyTcf was* upregulated** earliest, followed by local activation of Wnt genes. Among these, HyWnt3 and HyWnt11 appeared within 1.5 h of head removal, followed by HyWnt1, HyWnt9/10c, HyWnt16, and HyWnt7, indicating their role in the formation of the Hydra head organizer* (Hobmayer et al., 2001; Lengfeld et al., 2009; Philipp et al., 2009; Tursch et al., 2022).

    • The authors do not cite or reference at all the study by Cazet et al. which used iCRT14 along with RNAseq and ATACseq to probe the role of Wnt signaling during early regeneration. This is a major issue. Although I appreciate that the authors have done much longer time courses and that their data therefore add something to our understanding it will still be important to discuss here. For example, the authors show that Wnt3 is activated normally in iCRT14 animals. Is this also seen in the RNAseq from Cazet et al.*

    Answer:

    iCRT14 was used in Hydra regeneration experiments by Gufler et al (which we did cite) and Cazet et al, but the specific aspects of hypostome and tentacle regeneration were not addressed. Cazet et al. have analyzed HyWnt3expression after iCRT treatment during the first 12 hrs of regeneration. Our data show, in addition that HyWnt3 is not controlled by TCF-dependent transcription during Hydra head regeneration after apical cuts throughout the whole regeneration process including the morphogenesis state. Nevertheless, the other Wnt-genes, which are indicated in canonical Wnt-signalling are affected by iCRT14 also in our study.

    We have now included comparison of Cazet- and our data, we wrote:

    “HyWnt3 and Wnt9/10c expression are swiftly induced by injuries. When HyWnt3 and HyWnt9/10 activities are sustained, organizers can be formed, which induce ectopic heads when the original organizer tissue (the head) is removed (Cazet et al., 2021; Tursch et al., 2022).”

    The effect of iCRT14 had been analyzed in previous studies (Cazet et al., 2021; Gufler et al., 2018; Tursch et al., 2022). All showed b-catenin-dependency for down-regulation of head specific genes in foot regenerates at time points up to 12 hrs after head removal, including HyWnt3. They also stated a failure of head regeneration in the presence of iCRT14 but, in accordance with our study, did not reveal that HyWnt3 expression at future heads depended on b-catenin.* None of these studies analyzed the regeneration of tentacles and hypostomes separately and they did not report whether the regeneration of hypostomes 48 hrs after head removal occurred normally upon iCRT14 treatment. *

    • The visualizations used in Figure 3 are quite difficult to interpret and do to in all cases match descriptions in the text. The way the same type data is displayed in figure 5 so much nicer. It is also better to treat the same types of data in the same manner consistently throughout the paper. For Hes, for example, the authors state that there is a reduction although the data shows that this is very small and taking into account the 95% confidence interval does not seem to be significant. If this is the case then the positive control is not working in this experiment. This would be much clear if individual time points were compared like in figure 5 and statistical tests shown. The authors then state that Alx is not affected but there is actually a larger effect than what they deemed significant for Hes (the axes are notably different between these two and I think a more consistent axis would make the genes more comparable). Similarly, Gsc is described as being not affected at 8 hours but it appears again to change more that the positive control Hes. Given this I would call into question the validity of this dataset and/or the interpretation by the authors. A more thorough analysis including taking better into account statistical significance would go a long way to increasing confidence in this data. • The same issues in interpretation described for Figure 3 also apply to figure 4. The authors state that Wnt7 is affected less than Wnt1 and 3 but this is not evident in the figure and no comparative analysis is performed to confirm this. The same for Wnt 11 and 9/10c where what the authors description is very difficult to see in the figure. Sp5 is apparently upregulated, but this is not discussed. Again the axes are notably different making it even more difficult to compare between samples.* Answer____:

    We have now presented the data by simple scatter blots with significance information for every data point. This allows comparison between samples as requested by the reviewer. The GAMs were moved to the supplement. We believe that some readers may appreciate GAM-representation of the data because of the accessibility of the confidence interval over time.

    Concerning DAPT:

    “We now performed RT-qPCR analysis to compare gene expression dynamics of these genes during head regeneration 0, 8, 24, 36 and 48 hrs after head removal. Animals were either treated with 30 µM DAPT in 1% DMSO, or 1% DMSO as control for the respective time frames. Timepoint 0 was measured immediately after head removal. The results of these analyses revealed that HyHes expression was clearly inhibited by DAPT during the first 36 hrs after head removal (Fig. 3B), confirming previously published data which had indicated HyHes as a direct target for NICD (Münder et al., 2010). *HyAlx *expression levels were slightly up-regulated after 24 hrs, but later partially inhibited by DAPT (Fig. 3C). CnGsc expression under DAPT treatment initially (8hrs) was comparable to control levels, but then it was strongly inhibited (Fig. 3D). This goes along with the observed absence of organizer activity in regenerating Hydra tips (Münder et al., 2013). Interestingly, a similar result was seen for HySp5 expression, which was also normal at 8 hrs but was then inhibited by DAPT at later time points (Fig. 3E). HyKayak, while expression is normal after 8 hrs, was strongly overexpressed between 24 and 36 hrs of regeneration in DAPT-treated polyps in comparison to control regenerates (Fig. 3F).

    Concerning iCRT14

    Next, following the same procedure as described for DAPT, we compared the gene expression dynamics of iCRT14-treated regenerates with control regenerates. We found that the expression of HyWnt3 was not inhibited by iCRT14. In fact, it even appeared slightly up-regulated at the 8 hrs time point (Fig. 4A). Normal HyWnt3-expression at the end of the regeneration period was confirmed by in-situ hybridization for HyWnt3 as shown in Fig. 1D, indicating that HyWnt3 expression patterns and expression levels in ecto- and endodermal cells of the hypostome were faithfully regenerated (Fig. 4A). In contrast, HyAlx expression was completely abolished by iCRT14 (Fig. 4B), consistent with the observation that iCRT14-treated head regenerates did not regenerate any tentacles (Fig. 1A). HySp5 expression was not significantly affected by iCRT14 treatment at any time point (Fig. 4C).

    Furthermore, we found that *CnGsc *levels in iCRT14 remained similar to control regenerates up to 24 hrs, but were attenuated at later time points (Fig. 4D), very similar to the expression dynamics of the Notch-target gene HyHes (Fig. 4E). The expression of HyKayak was decreased at 8 hrs after head removal in the presence of iCRT14, but then increased above control levels after 48 hrs (Fig. 4F). There were no significant changes in the expression dynamics of HyBMP2/4 and HyBMP5/8b between iCRT14-treated regenerates and controls (Fig. 4G, H).”

    The precise number of biological replicates can be seen in the individual diagrams, they included for most genes three biological replicates, with always three technical replicates for each data point. Biological replicates were obtained over several years by different researchers. For some genes, we obtained very consistent data with high confidence in every experiment (e.g. HyWnt3, HyBMP4). We illustrate this in table 1, where three arrows indicate all such cases. With some genes we observed greater variation, which we interpret as no effect or a minor effect in table 1. Some of these variations may be explained by our observation of wave-like patterns in the expression dynamics. Therefore, we have included the following statement:

    “In addition, the gene expression dynamics for many of the analyzed genes appears in wave-like patterns in some experiments (see Figs S3 and S4). As we have only four time points measured, we cannot draw strong conclusions from these observations, except that some of the deviations in our data points (e.g. 48 hrs HyHes)”

    • In their description of figure 4 the authors completely omit to discuss the Cazet et al dataset which has the exact same early timepoints for iCRT14 treatment. This must be discussed and compared and any difference noted. * Answer:

    We included the iCRT14 results from Cazet et al., in our revised manuscript (see above).

    • End of page 11: The authors propose a model thereby the role of Notch in Wnt3 expression may be due to the presence of a repressor. However, I don't see any putative evidence at that stage. The authors also do not cite relevant work from both Cazet et al. and Tursch et al which show that Wnt3 is likely upregulated by bZIP TFs. In both these cases the authors show evidence of bZIP TF binding sites in the Wnt3 promoter along with other analyses. This is very relevantto the model presented by the authors here and must be discussed and compared. - * In particular the authors put forward HyKayak as an inhibitor of Wnt3 and this should be discussed along with the previous work.

    Answer:

    Tursch et al. 2022 did not claim that HyWnt3 is upregulated by bZiP TFs. They showed that HyWnt3 was strongly upregulated in a position-independent manner upon inhibition of the p38 or JNK (c-Jun N-terminal kinase) pathways (i.e., stress-induced MAPK pathways). This would rather support our hypothesis that HyKayak (AP-1 protein) might be a repressor of Wnt3-expression.

    Cazet et al have indicated that injury-responsive bZIP TFs are the most plausible regulators of canonical Wnt-signalling components during the early generic wound response. They identified CRE-elements, which can be bound by bZIP TFs, in the putative regulatory sequences of HyWnt3. However, they focused on the early stage of regeneration (0-12hpa), and showed that bZIP TFs, including jun, fos and creb are transiently upregulated at 3hpa and hypothesise that they could induce the upregulation of HyWnt3 at this stage as an injury response. We have to point out that the Hydra fos-homolog Hykayak, which our work is concerned with, is not identical with the fos-gene described in Cazet’s paper. In addition, the *Hykayak *gene was downregulated by Notch signalling during the morphogenesis state of regeneration (24-36 hrs), which is not the same stage investigated by Cazet et al. To avoid confusion, we have now included the Cazet-fos-sequence in our sequence comparison in Fig. S1 (fos_Cazet_HYDVU). Moreover, we have included more information about fos_Cazet in the context of a comparison with HyKayak.

    Different bZiP transcriptional factors (TFs) may have different effects on the expression of Wnt genes, and these effects are context-dependent. In previous research, Cazet et al. identified another Hydra fos gene (referred to as fos_cazet) and bZiP TF binding sites in the putative regulatory sequences of HyWnt3 and HyWnt9/10c. They showed that bZiP TF-genes, including Jun and fos, were transiently upregulated 3 hrs after amputation, therefore they hypothesized that bZiP TFs could induce TCF-independent upregulation of HyWnt3 during the early generic wound response (Cazet et al., 2021). However, in our study HyKayak expression continuously increased throughout the entire head regeneration process (Fig. 3E and 4E) including the morphogenesis stages (24-48 hrs post-amputation). Another study reported that inhibition of the JNK pathway (which disrupts formation of the AP-1 complex) resulted in upregulation of HyWnt3 expression in both, head and foot regenerates (Tursch et al., 2022). This result might support our hypothesis, but it only included the first 6 hours after amputation, similar to Cazet’s research. Therefore, it appears that HyKayak and fos_Cazet may have opposing roles in the regulation of Wnt-gene expression and are possibly activated by different signaling pathways depending on the stages of regeneration.

    • On page 12 the authors conclude based on gene expression in inhibitor treatment that there is a “change in complex composition of the two transcription factors.” This is something which would require biochemical evidence and I therefore suggest they remove this entirely. * Answer:

    we have removed this sentence

    • The authors use experiments in Craspedacusta to test their hypothesis of the role of Wnt and Notch signaling in Hydra. There is, in my opinion, an incorrect logic here. Regardless of the outcome, the roles of Wnt and Notch could conceivably be different in the two species and therefore testing hypothesis from one is not possible in the other. The authors should reframe their discussion of this to be more of a comparative framework. Moreover, the results do not necessarily indicate what the authors say. In Hydra Notch signaling is required to form the hypostome/mouth and this is not the case in Craspedacusta while Wnt signaling is required. The authors do not cite an important study from another Hydrozoan Hydractinia (Gahan at al.,2017). In that study the authors show that DAPT inhibits tentacles during regeneration but that the hypostome (or at least the arrangement of neurons and cnidocytes around the mouth) forms normally. This would indicate that in Hydractinia the process of head formation is different to in Hydra and would be congruent with what is shown here in Craspedacusta. This should be more thoroughly discussed, and all relevant literature cited.* Answer:

    We have concentrated our Craspedacusta work on Notch-signalling now. We only show that DAPT does not inhibit the regeneration of Craspedacusta heads. We have included new data showing that nevertheless it has an effect on the expression of hypothetical Notch target genes, but not on *CsWnt3 *(new Fig. 7). We have re-written our discussion accordingly and included the Hydractinia-work about Notch (Gahan2017). Although the Hydractinia paper lacks gene expression studies making it difficult to directly compare with the Hydra data, it supports our claim that Notch is required for regeneration of polyps with head and tentacles. We indeed do not know anything about Wnt-signalling in Craspedacusta. Our new results show that it is probably expressed in the head, because we observe very low levels of expression in the polyps after head removal, which increases considerably during regeneration of the head. This was included in the results:

    Results:

    “Finally, we investigated the expression of the Craspedacusta Wnt3-gene and its response to DAPT treatment during head regeneration. We observed low expression level of CsWnt3 after head removal (t=0), which dramatically increased as the head regenerated, suggesting that Wnt3 is expressed in the head of Craspedacusta polyps as it is in the head of other cnidarians including Hydra, Hydractinia and Nematostella (Hobmayer et al., 2000; Kusserow et al., 2005; Plickert et al., 2006). In accordance with having no effect on head regeneration, DAPT also did not inhibit CsWnt3 expression during this process in Craspedacusta. This is opposite to the situation in Hydra. If CsWnt3 would be involved in the Craspedacusta head regeneration, this could explain the failure of DAPT to interfere with this process”.

    Discussion part

    “Head regeneration also occurs in the colonial sea water hydrozoan Hydractinia. Colonies consist of stolons covering the substrate and connecting polyps, including feeding polyps, which have hypostomes and tentacles, and are capable of head regeneration, similar to Hydra polyps. Wnt3 is expressed at the tip of the head and by RNAi mediated knockdown it was shown that this gene is required for head regeneration (Duffy et al., 2010). In the presence of DAPT, Gahan et al observed that proper heads did not regenerate, similar to Hydra. However, they observed regeneration of the nerve ring around the hypostome indicating the possibility that hypostomes had been regenerated. Unfortunately, this study did not include gene expression data and therefore it is not clear whether Wnt3 expression was affected or not (Gahan et al., 2017).

    …..

    An interesting question was whether regeneration of cnidarian body parts, which are only composed of one module, also requires Notch-signalling. This is certainly true for the *Hydra *foot, which regenerates fine in the presence of DAPT (Käsbauer et al. 2007). Moreover, we tested head regeneration in Craspedacusta polyps, which do not have tentacles, and show that DAPT does not have an effect on this regeneration process. This corroborates our idea that Notch is required for regeneration in cnidarians, when this process involves two pattern forming processes to produce two independent structures, which are controlled by different signalling modules. This would be the case for the *Hydra *and for the Hydractinia heads, but not for Craspedacusta. ”

    …

    Moreover, we indicate at the end of our discussion that further studies about head regeneration in Craspedacusta and the genes involved would be desirable. We believe this would be beyond the scope of the current paper and we are working on a new Craspedacusta study.

    “Future studies on expression patterns of the genes that control formation of the Hydra head, including Sp5 and Alx in Craspedacusta could provide insights into the evolution of cnidarian body patterns. Sp5 and Alx appear to be conserved targets of Notch-signalling in the two cnidarians we have investigated. Wnt-3, while being inhibited by Notch-inhibition in Hydra head regenerates, is not a general target of Notch signalling. It was not affected by DAPT in our comparative transcriptome analysis (Moneer et al. 2021b) on uncut Hydra polyps, and it was also not affected by DAPT in regenerating heads of Craspedacusta.”

    • From reading the manuscript I do not fully understand the model the authors put forward. It is unclear what "coordinating two independent pattern forming systems" really means. It might be beneficial to make a schematic illustration of the model and how it goes wrong in both sets of inhibitor treatments. * Answer:

    We have edited the manuscript considerably and explained what we mean with the two pattern forming systems. It starts with the abstract:

    “Hydra head regeneration consists of two parts, hypostome/organizer and tentacle development.”

    …

    Thus, in accordance with regeneration of two head structures we find two signaling and gene expression modules with *HyWnt3 *and *HyBMP4 *part of a hypostome/organizer module, and BMP5/8, HyAlx and b-catenin part of a tentacle module. We conclude that Notch functions as an inhibitor of tentacle production in order to allow regeneration of hypostome/head organizer.

    …

    “Polyps of Craspedacusta do not have tentacles and thus, after head removal only regenerate a hypostome with a crescent of nematocytes around the mouth opening. This corroborates the idea that Notch-signaling mediates between two pattern forming processes during Hydra head regeneration”

    We have included the description of the organizer concept in the introduction, because we consider this relevant for our model:

    “The “organizer effect” entails a “harmonious interlocking of separate processes which makes up development”, or a side-by-side development of structures independently of each other (Spemann, 1935). In addition to inducing the formation of such structures, the organizer must ensure their patterning (Anderson and Stern, 2016). With reference to Hydra’s hydranth formation after head removal or transplantation, this involves the side-by side induction of hypostome tissue and tentacle tissue. Moreover, it includes the establishment of a regularly organized ring of tentacles with the hypostome doming up in the middle. The function of the Hydra“center of organization” would then be to pattern hypostome and tentacles and to allow for their harmonious re-formation after head removal”.

    In the discussion we integrate the organizer concept with the Gierer-Meinhardt reaction-diffusion models which still explain many aspects of Hydra development.

    “Is Notch part of the organizer? The organizer is defined as a piece of tissue with inductive and structuring capacity. Notch is expressed in all cells of Hydra polyps (Prexl et al., 2011) and overexpression of NICD does not induce second axes all over the Hydra body column (Pan et al., 2024), as seen with overexpression of stabilized b-catenin (Gee et al., 2010). Moreover, Notch functions differently during regeneration after apical and basal cuts. Phenotypically during head regeneration in DAPT, we clearly recognize a missing inhibition of tentacle tissue after apical cuts and missing inhibition of head induction after basal cuts (Pan et al., 2024). We would thus rather suggest that the organizer activity of Hydra tissue utilizes Notch-signaling as a mediator of inhibition. As our study of transgenic NICD overexpressing and knockdown polyps had suggested, the localization of Notch signaling cells depends on relative concentrations of Notch- and Notch-ligand proteins, which are established by gradients of signaling molecules that define the Hydra body axis (Pan et al., 2024; Sprinzak et al., 2010) . This is in very good agreement with a ”reaction-diffusion-model” provided by Alfred Gierer and Hans Meinhardt (Gierer and Meinhardt, 1972; Meinhardt and Gierer, 1974) suggesting a gradient of positional values across the Hydra body column. This gradient may determine the activities of two activation/inhibition systems, one for tentacles and one for the head. When the polyps regenerate new heads, Notch could provide inhibition for either system, depending on the position of the cut.

    We provide a new Fig. 8., which clearly illustrates the effects of DAPT and iCRT14 on hypostome and tentacle regeneration.

    Minor: • The abstract could be rewritten to have more of an introduction to the problem rather than jumping directly into results. It would also be beneficial if the abstract followed the logic of the paper.

    Answer: We agree and have re-written the abstract.

    • In Figure 3 and 4 it is not clear why they are divided into A and B. It appears that the categorization of genes into different groups lacks a clear rationale. This seems totally unnecessary. In addition, the order in which the genes are described in the text does not match what is seen in the figure making it confusing to follow. • In Figure 5 the authors use two different types of charts and I would stick with one. B is much better as it shows the individual data points as well as other information. I would use this throughout including in Figure 3 and 4. *

    __Answer: __

    We changed Fig. 3, 4 and 5 according to these comments and now present the data in one format over all three figures, in scatterplots (more detailed answer above).

    We are now describing the results in the order of the figures, with A and B omitted.

    Figure S3 is missing a description of panel C.

    In figure S3 it is not clear why the inhibitor was removed and not kept on throughout the experiment. Please discuss. __Answer: __

    Fig. S3 was removed.

    Figure S4 has no A or B in the figure, only in the legend. __Answer: __

    We have included A and B…

    *Reviewer #1 (Significance (Required)):

    Although some of the authors data appear to be novel I find the study makes only minor progress on the questions. In particular the authors do not properly cite the relevant literature and to put their manuscript into the correct context. The new model proposed by the authors is based entirely on qPCR data which is not thoroughly analyzed and are not strong enough in the absence of information about the spatial expression the genes they discuss. The proposal of HyKayak as a negative regulator of Wnt3 is interesting but the authors do not provide any solid direct evidence for this (ChIP, EMSA etc) and it is somewhat in disagreement with other models of bZIP function in the literature (which again are not discussed).*

    The manuscript is of limited general interest. It has a number of interesting observations which would be of interest to the Hydra community and the broader cnidarian community. The study lacks contextualization within a broader framework, whether it be in the context of regeneration or Wnt/Notch signaling. This limitation may narrow the overall interest in it.

    Answer:

    Our previous analysis of the effect of Notch on head regeneration in *Hydra *(Münder 2013) had suggested the inhibition model, which is part of Fig. 8. We show now that during head regeneration in Hydra formation of two structures is guided by different signaling/transcription modules, one using Wnt3 and BMP4, but not b-catenin; and one using BMP5/8 and b-catenin. We suggest that Notch functions as an inhibitor “of use” to the organizer when the “two-part” head structure is regenerated.

    We agree that our original manuscript was not well enough written to clearly put it into developmental context. We now focus the discussion of our work sharply on the organizer problem and think that the conclusions are of great general interest. In a simple view they suggest that the function of the *Hydra *head organizer is to allow harmonious development of head and tentacles, which we consider separate, and on a molecular basis independently regulated parts of the Hydra head. Notch signaling, in our interpretation, is an instrument of the organizer. Our comparison with *Craspedacusta *illustrates this idea. Craspedacusta only regenerates one head structure, which is possible in the absence of this instrument (also see reviewers 3 and 4).

    Concerning HyKayak, there is no disagreement with other authors as we analyze a fos-gene different from the one discussed by Cazet et al (see above). We have conducted a rescue experiments as suggested by reviewer 3 with the Kayak-inhibitor and with HyKayak shRNAi knockdown, however, rescue of the phenotype was not achieved although HyWnt3 was upregulated after DAPT treatment in the knockdown group. We attribute this to the very strong effect of DAPT. We have adjusted our hypothesis and only suggest that HyKayak could be a target for the Notch-induced repressor genes (e.g.HyHes). We mentioned this failed rescue in the manuscript (answer for see reviewer 3). Further experiments, e.g Chip/EMSA constitute a new project on the basis of these ideas and should be reserved for further studies of the Kayak-function in Hydra.

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

    *The study investigates the role of Notch and beta-catenin signaling in coordinating head regeneration in Hydra. It combines gene expression dynamics, inhibitor treatments, and comparisons with Craspedacusta polyps to propose a lateral inhibition model for Notch function during Hydra head regeneration, mediating between two pattern-forming systems.

    Three main concerns arise from this work:*

    • Lack of spatial expression data: The study proposes a model based on pattern-forming systems but falls short of providing direct spatial expression data for the genes under consideration in both control and treated scenarios. This gap weakens the empirical support for the proposed model.* Answer:

    The expression patterns for most of the presented genes including HyAlx and HyWnt3 in the presence and absence of DAPT have been published before (Münder 2013). Expression patterns for all other genes during regeneration (except Hykayak) are already known from literature. For Hykayak we have included expression data from Siebert et al (single cell transcriptome analysis) in the supplementary material. For iCRT14 treatment, we carried out a FISH-experiment and showed that HyWnt3 is expressed in the normal pattern at the hypostome. For further genes after DAPT and iCRT-treatment in situ hybridisation data are indeed lacking (e.g. BMP5/8). However, we have analyzed some very strongly downregulated regulated genes (e.g. HyAlx completely downregulated by iCRT14, all HyWnts and BMP2-4completely downregulated by DAPT) and for those in situ hybridisation could (1) be difficult due to low expression in treated samples and (2) may not be informative.

    • Clarity and relevance of Craspedacusta comparisons: The section discussing the regeneration in Craspedacusta polyps appears somewhat disjointed from the main narrative, with its contribution to the overarching story of Hydra regeneration remaining unclear. *

    Answer:

    We had not intended to explain gene expression during Craspedacusta head regeneration but wanted to prove our hypothesis that Notch is needed to allow side-by-side development of two newly arising structures, which use different signalling modules during head regeneration. That Notch is __not __needed for the regeneration of Craspedacusta, a polyp without tentacles, appears to strengthen our main hypothesis. In order to connect this point more clearly to the narrative we have included new data. We show that CsWnt3 expression lowers after head removal and rises when the head regenerates, indicating CsWnt3-expression in the head of Craspedacusta polyps. Moreover, we show now that Notch in Craspedacusta may have similar target genes as in Hydra (e.g. Sp5 and Alx), might also affect nematocyte differentiation as in Hydra, but does not inhibit Wnt3 expression. We also acknowledge that a precise understanding of the molecular pathways for head regeneration in *Craspedacusta *requires further work and have removed the results of iCRT14 treatment because of our lack of knowledge about the role of b-catenin in Craspedacusta patterning. Citations from our changed text are found in the answer to reviewer 1.

    • Accessibility of the text: The study's presentation, including its title, abstract, and main text, presents challenges in terms of clarity and accessibility, making it difficult for readers to follow and understand the research's scope, methodologies, and conclusions.*

    Answer:

    We agree and have completely re-written the abstract, and large parts of the introduction and discussion (also see above answer for reviewer 1).

    Reviewer #2 (Significance (Required)):

    In conclusion, while the study aims to advance our understanding of the complex signaling pathways governing Hydra head regeneration, it necessitates significant revisions. Enhancing the empirical evidence through detailed spatial patterning data, clarifying the comparative analysis with Craspedacusta polyps, and __refining the narrative __to improve accessibility are critical steps needed to solidify the study's contributions to the field.

    Answer:

    By including Kayak-expression data from Siebert et al and indicating the places of major expression of all analysed genes schematically in the Figs describing the qPCR data we revised our manuscript. We have added new data about Craspedacusta and believe that our re-written manuscript refines the narrative by focusing on the organizer (see answer to reviewer 1).

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

    Major comments:

    - In the abstract, the authors assert that their findings "indicate competing pathways for hypostome and regeneration." However, the nature of this competition and its resolution is not adequately elucidated within the manuscript. The term "competition" lacks context and clarity, leaving the reader without a clear understanding of what pathways are competing, for what, and how this competition is resolved during regeneration. Furthermore, this concept is not further explored or referenced throughout the remainder of the manuscript, leaving it somewhat disconnected from the main body of the research. It is recommended that the authors either revise the statement in the abstract to provide more clarity on the competing pathways and their implications for regeneration, or alternatively, if the authors believe there is sufficient evidence to support the claim of competing pathways, they should expand upon this point within the main body of the manuscript. Additional argumentation and evidence would be necessary to substantiate such a claim and provide a deeper understanding of the mechanisms underlying regeneration in Hydra.

    Answer:

    We agree and have removed any reference to “competing” pathways from the abstract and the main text.

    - The abstract makes a significant assertion regarding the mechanism by which Notch signaling impacts the expression of HyWnt3, suggesting that it operates by inhibiting HyKayak-mediated repression of HyWnt3 rather than directly activating transcription at the HyWnt3 promoter. This claim is central to the goals outlined in the study, which aim to elucidate the functioning of Notch signaling in HyWnt3 expression. To bolster this assertion, it would be prudent for the authors to conduct experiments demonstrating the mediating role of Kayak. Specifically, demonstrating that downregulation of Kayak through RNAi can rescue the DAPT-mediated downregulation of Wnt3 would provide strong support for the authors' claim. Additionally, while not strictly necessary, it would be beneficial to investigate whether chemical inhibition of Wnt can rescue the phenotype resulting from Kayak RNAi. Conducting and analyzing such experiments within a 2-3-month revision period should be feasible given that the authors already possess all necessary materials and have developed the required methods. These additional experiments would not only strengthen the evidence supporting the authors' claim but also provide further insights into the regulatory mechanisms at play in Notch signaling and HyWnt3 expression.

    Answer:

    We have conducted the suggested rescue experiments with the kayak-inhibitor, however, rescue was not achieved. We also tried rescue experiments by combining DAPT treatment and Kayak shRNA knockdown. HyWnt3 was slightly upregulated after DAPT treatment in the Kayak knockdown group but the phenotype could not be rescued. We are therefore now only state that HyKayak could be a target for the Notch-induced repressor genes (e.g.HyHes). We mentioned the failed rescue experiments in the manuscript:

    Results:

    *The up-regulation of HyKayak by DAPT suggests that HyKayak may serve as a potential target gene for Notch-regulated repressors including HyHes and CnGsc, potentially acting as a repressor of HyWnt3 gene transcription. *

    Discussion:

    We therefore suggest that the Hydra Fos-homolog HyKayak inhibits HyWnt3 expression and can be a target for a Notch-induced transcriptional repressor (like HyHes) in the regenerating Hydra head. Nevertheless, we were not able to rescue the DAPT-phenotype by inhibiting HyKayak, neither by using the inhibitor nor by shRNA-treatment, probably due to the strength of the DAPT effect. Therefore, we cannot exclude that Notch activates HyWnt3 directly, or that it represses unidentified Wnt-inhibitors through HyHes or CnGsc.

    - The usage of the term "lateral inhibition" in the title and abstract of the manuscript carries specific implications, as it is commonly associated with distinct mechanisms in the context of Notch signaling and reaction-diffusion systems. Notably, in the Notch signaling context, lateral inhibition typically refers to the amplification of small differences between neighboring cells through direct interactions, facilitated by the limitations of Notch signaling to immediate neighbors. Conversely, in reaction-diffusion systems, such as the Gierer-Meinhardt model, lateral inhibition describes long-range inhibition associated with pattern formation.

    Given this discrepancy, it is crucial for the authors to clarify their interpretation of "lateral inhibition" to avoid ambiguity and ensure accurate understanding. If they are referring to Notch-specific lateral inhibition, they should provide evidence of adjacent localization of Notch and Delta cells to support their argument. Alternatively, if they are invoking the concept of long-range inhibition described by the Gierer-Meinhardt model, they must explain how a membrane-tethered ligand like Notch can exert effects beyond one cell diameter from the signaling center.

    Regardless of the interpretation chosen by the authors, addressing this clarification will have significant implications for the subsequent treatment of their arguments. Depending on their chosen interpretation, experimental demonstrations may be necessary to substantiate their claims, which could be laborious and time-consuming. However, such demonstrations are essential for establishing the validity of the term "lateral inhibition" as used in the title and abstract of the manuscript.*

    Answer:

    We agree with the reviewer concerning the term “lateral inhibition” and have now removed it. Instead, we have emphasized that our data clearly show during apical regeneration a Notch-mediated inhibition of tentacle tissue formation. We also discuss data from our most recent publication (Pan 2024) showing that this is the opposite at basal cuts, where the loss of Notch function leads to the regeneration of two heads. We then discuss this in the context of the Gierer-Meinhardt Model and in the context of the organizer (also see above in answer to reviewer 1):

    It is true that it is difficult to reconcile the long-range signaling processes, on which the Gierer-Meinhardt model is based with the cell-cell interactions mediated by Notch-signaling. We have now published a mathematical model to explain our understanding of this for the role of Notch during budding and in steady state animals (Pan2024), which is based on work by Sprinzak et al 2010. For head regeneration, we do not have such a model yet. Here we are looking at expression patterns changing over time. Therefore, we assume waves of gene expression, relying on the autoinhibitory function of the HyHes-repressor. This is included in the discussion:

    In addition, the gene expression dynamics for many of the analyzed genes appears in wave-like patterns in some experiments (see Figs S3 and S4). As we have only four time points measured, we cannot draw strong conclusions from these observations, except that some of the deviations in our data points (e.g. 48 hrs HyHes) might be caused by oscillations. Nevertheless, we propose that the dynamic development of gene expression patterns over the time course of regeneration hint at a wave like expression of Notch-target genes (e.g. HyAlx, (Münder et al., 2013; Smith et al., 2000)). Hes-genes have been implicated in mediating waves of gene expression, e.g. during segmentation and as part of the circadian clock (Kageyama et al., 2007). This property is due to the capability of Hes-proteins to inhibit their own promoter. Future models for head regeneration in Hydra should consider the function of Notch to inhibit either module of the regeneration process and the potential for the Notch/Hes system to cause waves of gene expression. Such waves intuitively seem necessary to change the gene expression patterns underlying morphogenesis during the time course of head regeneration.

    - The utilization of Craspedacusta as a comparative model in the argumentation of the manuscript appears somewhat unclear. The authors posit that Notch is essential for organizer emergence in Hydra, while Wnt is not necessary, as indicated by the observed effects of iCRT14 beta-catenin/TCF inhibition. However, in Craspedacusta, which lacks tentacles but possesses an organizer, one might anticipate a conserved requirement for organizer formation but not tentacle development. Therefore, it would be reasonable to expect that Craspedacusta would still form an organizer under iCRT14 treatment but would not depend on Notch signaling, as the necessity to separate tentacle formation from organizer formation is absent. The authors' observation that Craspedacusta fails to form an organizer under iCRT14 treatment partially aligns with these expectations. However, the complexity of the results suggests a need for a deeper understanding of the involvement of different pathways in Craspedacusta. Before applying inhibitors, it would be crucial to elucidate the spatiotemporal differences in the expression of relevant Wnt and Notch pathway components between Hydra and Craspedacusta. This knowledge would provide valuable insights into the specific roles of these pathways in organizer formation and tentacle development in both species, helping to clarify the observed differences in response to iCRT14 treatment. Additionally, considering the possibility of differential sensitivity to iCRT14 (see comment below) between Hydra and Craspedacusta would be essential for accurately interpreting the results and drawing meaningful conclusions regarding the involvement of Notch and Wnt signaling pathways in these processes.

    * *

    Answer:

    We have clarified in our re-written manuscript that the organizer functions in Hydra heads and head regeneration to harmonize the development of two independent structures (see answer for reviewer 1) and that Notch-signalling is an instrument to achieve this. Craspedacusta polyps do not have tentacles, thus we do not see two independent structures. Correspondingly, we see that they do not need Notch-signaling. We do not know whether they have organizer tissue, because they are too small to perform transplantation experiments. Similarly, in situ hybridisation experiments to look for CsWnt expression are hard to envisage. What we have now done during the revision of this paper are RT-qPCR experiments to follow the expression of CsWnt3 after head removal until a new head is formed. This indicated the localization of CsWnt3 expression in the head (citations in response to reviewer 1).

    We agree that the role of Wnt/b-catenin for Craspedacusta cannot be sufficiently described with our iCRT14 experiment and therefore removed it. To strengthen the DAPT data, we also examined Craspedacusta homologs of the Hydra Notch-target genes that we had previously identified (Moneer2021). We found that expression of CsSp5 and *CsAlx *were inhibited by DAPT. This was also true for the nematocyte gene *NOWA *(see new Fig. 7). In Hydra, DAPT blocks one important differentiation step of nematocytes and therefore the expression of all genes expressed in differentiating capsule precursors, including NOWA is inhibited, while the number of mature capsules does not change. To see the same DAPT effect on NOWA-expression in *Craspedacusta *reassured us that DAPT had entered the animals and might also have a similar effect on nematocytes as in Hydra.

    *Minor comments

    • The concentration-dependent effects of iCRT14 on beta-catenin signaling, as demonstrated by Gufler et al. 2018, suggest that the efficacy of inhibition may vary depending on the concentration used. Specifically, Gufler et al. found that a concentration of 10µM was sufficient for efficient inhibition of beta-catenin signaling. However, in the current study, the authors utilized a concentration of 5µM of iCRT14. Given the central role of the observed effects, particularly the persistence of Wnt3 expression, in the argumentation of the manuscript, it is plausible that these effects could be attributed to partial inhibition resulting from the lower concentration of iCRT14 used in the study. To address this potential limitation, the authors could consider conducting a quick examination of the effects of 10µM iCRT14 or utilizing other inhibitors of beta-catenin/TCF interaction, such as iCRT3. By comparing the effects of different concentrations or alternative inhibitors, the authors could ascertain whether the observed effects are indeed attributable to partial inhibition from 5µM iCRT14, or if they persist despite higher concentrations or alternative inhibitors. This additional experimentation would provide valuable insights into the specificity and efficacy of the inhibition and strengthen the validity of the conclusions drawn regarding the role of beta-catenin signaling in the observed phenomena.*

    Answer:

    The iCRT14 concentration was adjusted to 5 µM because the initial 10µM proved to be too toxic. 5µM also produced the same phenotypes and results as seen before. Cazet et al. also used 5 µM iCRT14 in their study.

    - The use of Generalized Additive Models (GAMs) in Figures 3 and 4 to present the time series qPCR results may introduce some challenges in interpretation due to the potential for distortion of values at specific time points based on neighboring ones. Given the relatively low time resolution of the data, this approach could lead to a distorted depiction of the temporal dynamics. For instance, in Figure 3B, where Wnt3 peaks at 10 hours, the absence of measurements between 8 and 24 hours introduces uncertainty regarding the accuracy and reliability of this peak at 10 hours.

    To address these concerns and enhance clarity, it may be advisable for the authors to consider presenting the data using simple boxplots instead of GAMs. Boxplots provide a more straightforward visualization of the distribution of data at each time point, allowing for a clearer interpretation of trends and fluctuations over time. This approach would mitigate the potential for distortion introduced by GAMs and provide a more accurate representation of the temporal dynamics observed in the qPCR results*

    Answer:

    We agree and have changed the data representation to simple scatterplots (see answers for reviewer 1).

    - The comparison of the effects of iCRT14 versus DAPT treatments would benefit from having consistent gene expression data across both treatments. However, in Figure 4A, there are fewer genes tested compared to Figure 3A, with Hes and Kayak omitted. While the authors interpretation suggests that these genes may not change after iCRT14 treatment due to their upstream position in the signaling pathways, it is essential to empirically demonstrate this relationship, as it is central to the conclusions drawn. To address this gap in the analysis, it would be valuable for the authors to provide a time series of differential expression for Hes and Kayak following iCRT14 treatment.

    Answer:

    We have provided a time series for expression of HyHes and HyKayak in responses to iCRT14 treatment during regeneration (see Fig.4).

    “We found that the expression the Notch-target gene HyHes remained similar to control regenerates up to 24 hrs, but then was attenuated (Fig. 4A), possibly due to failure of tentacle boundary formation, the tissue where HyHes is strongly expressed…The expression of HyKayak was decreased at 8 hrs after head removal in the presence of iCRT14, came back to normal up to 36 hrs and was suddenly increased after 48 hrs (Fig. 4E), correlating with inhibition of the HyHes repressor. There were no significant changes in the expression dynamics of HyBMP2/4 and HyBMP5/8b between iCRT14-treated regenerates and controls (Fig. 4F, G).”

    - The analysis of the impact of chemical inhibition of Notch and Wnt signaling in Figure 7 schematic highlights changes in spatial expression patterns of the target genes. However, the interpretation of their impact primarily relies on qPCR data. As evident from Figure 7, when Notch is inhibited, it is anticipated that Kayak expression will shift from the area of the tentacles to the tip. This spatial shift in expression patterns is a critical aspect of the authors' arguments, especially considering the centrality of Kayak in their findings. Notably, similar spatial expression patterns have been demonstrated for Alx using FISH in Pan et al., available on BioRxiv. Given the importance of Kayak in the presented arguments, it is advisable to also investigate its spatial expression patterns using techniques such as FISH.

    Answer:

    We have, instead of FISH-experiments, included expression data for HyKayak from Siebert et al 2019 (single cell transcriptome data) in Fig. S1D, which show its expression in head- and battery cells (tentacle cells). This is similar to HyAlx. Therefore, Kayak-FISH would be expected to reveal expression of the gene at the tip of the regenerate the whole time, similar to HyAlx, because tentacle gene inhibition or patterning does not occur (see Münder 2013). Due to the failure of our rescue experiment to demonstrate the function of kayak we have omitted kayak from Fig. 8 and only mention in the discussion that it could be a target for Notch activated transcriptional repressors, like HyHes or CnGsc.

    Reviewer #3 (Significance (Required)):

    *The paper introduces novelties to the field of regeneration and developmental biology by leveraging Craspedacusta polyp as a novel model system for investigating the evolutionary and developmental dynamics of tentacles. In doing so, it sheds new light on the intricate mechanisms underlying tentacle formation and patterning. Furthermore, the study implicates Kayak in the regulation of Wnt3, adding a fresh perspective to our understanding of the molecular pathways governing Hydra regeneration. Notably, the results of the research challenge the prevailing notion of autoregulation of Wnt3, which has long been considered fundamental to organizer formation in Hydra. While these findings offer intriguing insights, further investigation will be crucial to conclusively ascertain the validity of this assertion. *

    Despite the clarity of the data presented, the interpretation and integration of these findings in the manuscript are lacking. The narrative at times feels disjointed, with different storylines loosely connected. While the findings are intriguing and merit publication, a substantial revision of the manuscript is necessary to provide a more coherent and illuminating interpretation of the results. *The implications of this research extend beyond the specific confines of Craspedacusta polyp and Hydra biology. It holds significant relevance for both the Hydra biology community and the broader field of Notch signaling research. *

    By highlighting the pivotal role of Notch signaling in regeneration and patterning within Hydra, the study enriches our comprehension of this model organism and its evolutionary adaptations. Moreover, it provides a valuable lens through which the evolution of Notch signalling cascades can be examined. This interdisciplinary approach underscores the interconnectedness of diverse biological systems and underscores the importance of exploring novel model organisms to unravel the complexities of evolution and development.

    Answer:

    We have edited the manuscript considerably and re-written the introduction and the discussion parts. We are focusing on integrating this work with the organizer concept in developmental biology, and on the Gierer-Meinhardt-model, and point out that Notch-signaling is required for the development of two head structures by inhibiting the development of either one during head regeneration, which is necessary to enable the development of the other one. Which one is inhibited depends on the positional value of the tissue where the cut occurs. Craspedacusta polyps do only have one structure. We suggest that this is why head regeneration does not require Notch-signalling in Craspedacusta. In contrast, as we have included in our discussion now, *Hydractinia *polyps, again with head/mouth and tentacles, require Notch-signaling for head regeneration (according to Gahan 2019), see also answers for reviewers 1 and 2.

    Reviewer #4 (Evidence, reproducibility and clarity (Required)):

    Major comments:

    The conclusions from the experiments are drawn accurately, not overstating the results. The main conclusion, that in Hydra Notch pathway mediates between two patterning modules, hypostome and tentacle forming modules, is supported by in situ hybridization and qPCR analyses of hypostome and tentacle specific genes.

    OPTIONAL. Authors hypothesize, that Notch maintains expression of Wnt3 vie its targets, transcriptional repressors Goosecoid or Hes, which halt the expression of Wnt3 repressor HyKayak. Epistatic relationships between Notch, Goosecoid or Hes and HyKayak could be tested, first, by combining pharmacological inhibition of Notch by DAPT with shRNA-mediated knockdown and second, in double knockdowns generated by electroporating shRNAs for two genes simultaneously. If the proposed in the pathway relationships are correct the repressive effect of DAPT treatment on an organizer regeneration should be rescued in HyKayak shRNA-mediated knockdown. Regeneration of an organizer also should occur in Notch/HyKayak and Goosecoid (Hes)/HyKayak shRNA-mediated double knockdowns. Electroporation of shRNAs for multiple genes is an effective and quick way to generate double and triple knockdowns. The proposed experiments will much strengthen the conclusions drawn from this study. Given that the authors have successfully used shRNA-mediated technique to generate HyKayak knockdown animals, they should be able to complete the proposed experiments within in a couple of months. Answer:

    We very much like the suggested strategy to probe the regeneration pathways by shRNA-mediated knockdown experiments- this will be a basis for future investigations.

    We conducted the suggested rescue experiment by combining the DAPT treatment and Kayak shRNA knockdown. HyWnt3 was slightly upregulation after DAPT treatment in the Kayak knockdown group. However, this upregulation did not rescue the organizer’s regeneration. We think that the effect of DAPT is too strong. We have included this in the discussion of our results (see answer for reviewer 2).

    • The data are presented in a logical and clear manner. The paper is easy to read, and the conclusions are explicit for each experimental section. The methodology is described in detail and should be easy to reproduce.*
    • All experiments are done with multiple biological and technical replicates. However, the description of statistical analysis used in each case is missing, p values and error bars are missing in Fig. 2B and Fig. S4. Author should add this information in the main text or in the figure legends.*

    Answer:

    The statistical information was now added in the methods section.

    Minor comments:

    • Fig. 1E. It would be more convincing to present tentacle and hypostome regeneration data separately, comparing hypostome regeneration in treated animals with DMSO control, and in a separate analysis comparing tentacle regeneration with control. Provide the description of statistical method, p values and error bars. If authors prefer to stick to the current way of presenting they should also provide description of statistical analysis used and statistical data.*

    Answer:

    We changed the representation in Fig. 1E. We now use scatter plots in the main text with p-values added, and explained the statistics of the GAM representation in the supplementary material.

    • Results, section 4 Kayak. Authors use T5424 inhibitor to block the potential interactions between HyKayak with HyJun. The resulted increase in Wnt3 expression measured by qPCR clearly supports the idea of HyKayak being a repressor of Wnt3. However, authors are going further and present the phenotype of T5424 treatment, shortening of the tentacles. Many factors can influence the length of the tentacles. For example, shortening of tentacles is a strong indication of poisoning or animal being in general unwell. At a concentration double of the one used in the experiment T5424 causes a disintegration of the animals (Fig. 3S). It would be more convincing if the authors could provide an in situ hybridization image showing an expansion of Wnt3 expression domain down the hypostome. This is the result one would expect from the inhibition of HyKayak which, according to the proposed mechanism, restricts Wnt3 spatial expression to the most apical portion of the regenerating tip. Alternatively, authors could try to see if T5424 rescues the inhibition of an organizer formation resulted by DAPT treatment. The latter experiment might be difficult to perform due to a possible toxic effect of multidrug treatment. I suggest that authors either include the proposed experiments or leave the results of the Fig S3 out.*

    Answer:

    According to this suggestion we have removed the phenotypes of polyps after treatment with T5424.

    • Results, section 3.2, paragraph 4. 'This also applies for the suggested Hydra organizer gene CnGsc, and BMP2/4 (Broun, Sokol et al. 1999). Please, insert the citation for BMP2/4.*

    Answer:

    We inserted the citation for BMP2-4 (Watanabe 2014).

    Reviewer #4 (Significance (Required)):

    *Significance:

    The current study is a continuation of the author's previous work where they have characterized Notch pathway in Hydra and showed its role in the regeneration of an organizer and patterning of Hydra head. Here, they present the study of Notch pathway in the context of b-catenin pathway, a pathway that has been shown to be essential for the axis induction and patterning in Hydra. The authors challenge this dogma and show, that during head regeneration b-catenin transcriptional activity is not required either to maintain the expression of wnt3 nor to acquire an inductive activity of the regenerating organizer. Second, they show, that transcriptional fos-related factor Kayak is negatively regulated by Notch-signaling and, in turn, represses transcription of Wnt3. Based on those findings authors propose a function of the Notch pathway in Hydra head regeneration, particularly in spatial separation of the hypostome/organizer module from the tentacle module. The role of Notch pathway in lateral inhibition is well documented in bilaterians. However, in Cnidaria, a sister group to Bilateria, the function of Notch was so far restricted to neurogenesis. This study is very important for our understanding of the evolution of morphogenesis as it shows the ancient role that the Notch pathway is playing in axial patterning, possibly, through lateral inhibition.

    This study can be of a great interest to both researchers specializing in cnidarian development and to a broader audience interested in the evolution of morphogenesis.*

  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 #4

    Evidence, reproducibility and clarity

    This is a very nice study exploring the function of Notch pathway in Hydra, a member of early Metazoa Cnidaria. The main conclusions of the study are:

    1. -catenin pathway is required not for regeneration of Hydra hypostome/ the organizer, as previously thought, but rather for regeneration of tentacles and correct patterning of Hydra head.
    2. During head regeneration Notch pathway, possibly, through lateral inhibition, blocks tentacle fate at the most apical region of the regenerating tip, allowing a hypostome/an organizer to develop. This might occur via the targets of Notch pathway, transcriptional repressors Goosecoid and Hes.
    3. During head regeneration Notch pathway is required to maintain the expression of Hydra organizer gene Wnt3 as well as other canonical wnt genes. This occurs, possibly, through repression of HyKayak, Hydra homologue of a transcriptional fos-related factor Kayak, that, in turn, represses Wnt3.

    Major comments:

    1. The conclusions from the experiments are drawn accurately, not overstating the results. The main conclusion, that in Hydra Notch pathway mediates between two patterning modules, hypostome and tentacle forming modules, is supported by in situ hybridization and qPCR analyses of hypostome and tentacle specific genes.
    2. OPTIONAL. Authors hypothesize, that Notch maintains expression of Wnt3 vie its targets, transcriptional repressors Goosecoid or Hes, which halt the expression of Wnt3 repressor HyKayak. Epistatic relationships between Notch, Goosecoid or Hes and HyKayak could be tested, first, by combining pharmacological inhibition of Notch by DAPT with shRNA-mediated knockdown and, second, in double knockdowns generated by electroporating shRNAs for two genes simultaneously. If the proposed in the pathway relationships are correct the repressive effect of DAPT treatment on an organizer regeneration should be rescued in HyKayak shRNA-mediated knockdown. Regeneration of an organizer also should occur in Notch/HyKayak and Goosecoid(Hes)/HyKayak shRNA-mediated double knockdowns. Electroporation of shRNAs for multiple genes is an effective and quick way to generate double and triple knockdowns. The proposed experiments will much strengthen the conclusions drawn from this study. Given that the authors have successfully used shRNA-mediated technique to generate HyKayak knockdown animals, they should be able to complete the proposed experiments within in a couple of months.
    3. The data are presented in a logical and clear manner. The paper is easy to read, and the conclusions are explicit for each experimental section. The methodology is described in detail and should be easy to reproduce.
    4. All experiments are done with multiple biological and technical replicates. However, the description of statistical analysis used in each case is missing, p values and error bars are missing in Fig. 2B and Fig. S4. Author should add this information in the main text or in the figure legends.

    Minor comments:

    1. Fig. 1E. It would be more convincing to present tentacle and hypostome regeneration data separately, comparing hypostome regeneration in treated animals with DMSO control, and in a separate analysis comparing tentacle regeneration with control. Provide the description of statistical method, p values and error bars. If authors prefer to stick to the current way of presenting they should also provide description of statistical analysis used and statistical data.
    2. Results, section 4 Kayak. Authors use T5424 inhibitor to block the potential interactions between HyKayak with HyJun. The resulted increase in Wnt3 expression measured by qPCR clearly supports the idea of HyKayak being a repressor of Wnt3. However, authors are going further and present the phenotype of T5424 treatment, shortening of the tentacles. Many factors can influence the length of the tentacles. For example, shortening of tentacles is a strong indication of poisoning or animal being in general unwell. At a concentration double of the one used in the experiment T5424 causes a disintegration of the animals (Fig. 3S). It would be more convincing if the authors could provide an in situ hybridization image showing an expansion of Wnt3 expression domain down the hypostome. This is the result one would expect from the inhibition of HyKayak which, according to the proposed mechanism, restricts Wnt3 spatial expression to the most apical portion of the regenerating tip. Alternatively, authors could try to see if T5424 rescues the inhibition of an organizer formation resulted by DAPT treatment. The latter experiment might be difficult to perform due to a possible toxic effect of multidrug treatment. I suggest that authors either include the proposed experiments or leave the results of the Fig S3 out.
    3. Results, section 3.2, paragraph 4. 'This also applies for the suggested Hydra organizer gene CnGsc, and BMP2/4 (Broun, Sokol et al. 1999). Please, insert the citation for BMP2/4.

    Significance

    The current study is a continuation of the author's previous work where they have characterized Notch pathway in Hydra and showed its role in the regeneration of an organizer and patterning of Hydra head. Here, they present the study of Notch pathway in the context of -catenin pathway, a pathway that has been shown to be essential for the axis induction and patterning in Hydra. The authors challenge this dogma and show, that during head regeneration -catenin transcriptional activity is not required either to maintain the expression of wnt3 nor to acquire an inductive activity of the regenerating organizer. Second, they show, that transcriptional fos-related factor Kayak is negatively regulated by Notch signaling and, in turn, represses transcription of Wnt3. Based on those findings authors propose a function of the Notch pathway in Hydra head regeneration, particularly in spatial separation of the hypostome/organizer module from the tentacle module. The role of Notch pathway in lateral inhibition is well documented in bilaterians. However, in Cnidaria, a sister group to Bilateria, the function of Notch was so far restricted to neurogenesis. This study is very important for our understanding of the evolution of morphogenesis as it shows the ancient role that the Notch pathway is playing in axial patterning, possibly, through lateral inhibition.

    This study can be of a great interest to both researchers specializing in cnidarian development and to a broader audience interested in the evolution of morphogenesis.

    The reviewer's field of expertise includes cnidarian development, axial patterning and morphogenesis in Hydra, biochemical pathways in Hydra axial patterning, Hippo pathway regulation and tissue patterning in multiple organisms

  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 #3

    Evidence, reproducibility and clarity

    Summary:

    In the reviewed study by Steichele et al., the authors investigate the roles of Notch and Wnt signaling pathways in the regeneration process of Hydra hypostome and tentacles. They observe that treatment with DAPT prevents the emergence of an organizer and hypostome, whereas inhibition of beta-catenin/TCF interaction does not affect organizer formation but hinders tentacle development. The authors delve into the molecular mechanisms underlying these observations, examining the impact of different treatments on the expression of genes crucial for organizer and tentacle emergence. Notably, they identify Kayak as a target of Notch signaling, which is upregulated following DAPT treatment. Furthermore, RNAi against Kayak results in the overexpression of Wnt3, suggesting an inhibitory role of Kayak on Wnt3 expression. This inhibitory effect is corroborated using the Fos/Jun inhibitor T5224. In their investigation, the authors compare the effects of Notch and Wnt signaling inhibition using a polyp species, Craspedacusta, which possesses an organizer but lacks tentacles. They find that while Notch inhibition does not affect this polyp, inhibition of canonical Wnt signaling does.

    Major comments:

    • In the abstract, the authors assert that their findings "indicate competing pathways for hypostome and regeneration." However, the nature of this competition and its resolution is not adequately elucidated within the manuscript. The term "competition" lacks context and clarity, leaving the reader without a clear understanding of what pathways are competing, for what, and how this competition is resolved during regeneration. Furthermore, this concept is not further explored or referenced throughout the remainder of the manuscript, leaving it somewhat disconnected from the main body of the research.

    It is recommended that the authors either revise the statement in the abstract to provide more clarity on the competing pathways and their implications for regeneration, or alternatively, if the authors believe there is sufficient evidence to support the claim of competing pathways, they should expand upon this point within the main body of the manuscript. Additional argumentation and evidence would be necessary to substantiate such a claim and provide a deeper understanding of the mechanisms underlying regeneration in Hydra.

    • The abstract makes a significant assertion regarding the mechanism by which Notch signaling impacts the expression of HyWnt3, suggesting that it operates by inhibiting HyKayak-mediated repression of HyWnt3 rather than directly activating transcription at the HyWnt3 promoter. This claim is central to the goals outlined in the study, which aim to elucidate the functioning of Notch signaling in HyWnt3 expression. To bolster this assertion, it would be prudent for the authors to conduct experiments demonstrating the mediating role of Kayak. Specifically, demonstrating that downregulation of Kayak through RNAi can rescue the DAPT-mediated downregulation of Wnt3 would provide strong support for the authors' claim. Additionally, while not strictly necessary, it would be beneficial to investigate whether chemical inhibition of Wnt can rescue the phenotype resulting from Kayak RNAi. Conducting and analyzing such experiments within a 2-3-month revision period should be feasible given that the authors already possess all necessary materials and have developed the required methods. These additional experiments would not only strengthen the evidence supporting the authors' claim but also provide further insights into the regulatory mechanisms at play in Notch signaling and HyWnt3 expression.
    • The usage of the term "lateral inhibition" in the title and abstract of the manuscript carries specific implications, as it is commonly associated with distinct mechanisms in the context of Notch signaling and reaction-diffusion systems. Notably, in the Notch signaling context, lateral inhibition typically refers to the amplification of small differences between neighboring cells through direct interactions, facilitated by the limitations of Notch signaling to immediate neighbors. Conversely, in reaction-diffusion systems, such as the Gierer-Meinhardt model, lateral inhibition describes long-range inhibition associated with pattern formation.

    Given this discrepancy, it is crucial for the authors to clarify their interpretation of "lateral inhibition" to avoid ambiguity and ensure accurate understanding. If they are referring to Notch-specific lateral inhibition, they should provide evidence of adjacent localization of Notch and Delta cells to support their argument. Alternatively, if they are invoking the concept of long-range inhibition described by the Gierer-Meinhardt model, they must explain how a membrane-tethered ligand like Notch can exert effects beyond one cell diameter from the signaling center.

    Regardless of the interpretation chosen by the authors, addressing this clarification will have significant implications for the subsequent treatment of their arguments. Depending on their chosen interpretation, experimental demonstrations may be necessary to substantiate their claims, which could be laborious and time-consuming. However, such demonstrations are essential for establishing the validity of the term "lateral inhibition" as used in the title and abstract of the manuscript.

    • The utilization of Craspedacusta as a comparative model in the argumentation of the manuscript appears somewhat unclear. The authors posit that Notch is essential for organizer emergence in Hydra, while Wnt is not necessary, as indicated by the observed effects of iCRT14 beta-catenin/TCF inhibition. However, in Craspedacusta, which lacks tentacles but possesses an organizer, one might anticipate a conserved requirement for organizer formation but not tentacle development. Therefore, it would be reasonable to expect that Craspedacusta would still form an organizer under iCRT14 treatment but would not depend on Notch signaling, as the necessity to separate tentacle formation from organizer formation is absent.

    The authors' observation that Craspedacusta fails to form an organizer under iCRT14 treatment partially aligns with these expectations. However, the complexity of the results suggests a need for a deeper understanding of the involvement of different pathways in Craspedacusta. Before applying inhibitors, it would be crucial to elucidate the spatiotemporal differences in the expression of relevant Wnt and Notch pathway components between Hydra and Craspedacusta. This knowledge would provide valuable insights into the specific roles of these pathways in organizer formation and tentacle development in both species, helping to clarify the observed differences in response to iCRT14 treatment. Additionally, considering the possibility of differential sensitivity to iCRT14 (see comment below) between Hydra and Craspedacusta would be essential for accurately interpreting the results and drawing meaningful conclusions regarding the involvement of Notch and Wnt signaling pathways in these processes.

    Minor comments

    • The concentration-dependent effects of iCRT14 on beta-catenin signaling, as demonstrated by Gufler et al. 2018, suggest that the efficacy of inhibition may vary depending on the concentration used. Specifically, Gufler et al. found that a concentration of 10µM was sufficient for efficient inhibition of beta-catenin signaling. However, in the current study, the authors utilized a concentration of 5µM of iCRT14. Given the central role of the observed effects, particularly the persistence of Wnt3 expression, in the argumentation of the manuscript, it is plausible that these effects could be attributed to partial inhibition resulting from the lower concentration of iCRT14 used in the study. To address this potential limitation, the authors could consider conducting a quick examination of the effects of 10µM iCRT14 or utilizing other inhibitors of beta-catenin/TCF interaction, such as iCRT3. By comparing the effects of different concentrations or alternative inhibitors, the authors could ascertain whether the observed effects are indeed attributable to partial inhibition from 5µM iCRT14, or if they persist despite higher concentrations or alternative inhibitors. This additional experimentation would provide valuable insights into the specificity and efficacy of the inhibition and strengthen the validity of the conclusions drawn regarding the role of beta-catenin signaling in the observed phenomena.
    • The use of Generalized Additive Models (GAMs) in Figures 3 and 4 to present the time series qPCR results may introduce some challenges in interpretation due to the potential for distortion of values at specific time points based on neighboring ones. Given the relatively low time resolution of the data, this approach could lead to a distorted depiction of the temporal dynamics. For instance, in Figure 3B, where Wnt3 peaks at 10 hours, the absence of measurements between 8 and 24 hours introduces uncertainty regarding the accuracy and reliability of this peak at 10 hours.

    To address these concerns and enhance clarity, it may be advisable for the authors to consider presenting the data using simple boxplots instead of GAMs. Boxplots provide a more straightforward visualization of the distribution of data at each time point, allowing for a clearer interpretation of trends and fluctuations over time. This approach would mitigate the potential for distortion introduced by GAMs and provide a more accurate representation of the temporal dynamics observed in the qPCR results

    • The comparison of the effects of iCRT14 versus DAPT treatments would benefit from having consistent gene expression data across both treatments. However, in Figure 4A, there are fewer genes tested compared to Figure 3A, with Hes and Kayak omitted. While the authors interpretation suggests that these genes may not change after iCRT14 treatment due to their upstream position in the signaling pathways, it is essential to empirically demonstrate this relationship, as it is central to the conclusions drawn. To address this gap in the analysis, it would be valuable for the authors to provide a time series of differential expression for Hes and Kayak following iCRT14 treatment.
    • The analysis of the impact of chemical inhibition of Notch and Wnt signaling in Figure 7 schematic highlights changes in spatial expression patterns of the target genes. However, the interpretation of their impact primarily relies on qPCR data. As evident from Figure 7, when Notch is inhibited, it is anticipated that Kayak expression will shift from the area of the tentacles to the tip. This spatial shift in expression patterns is a critical aspect of the authors' arguments, especially considering the centrality of Kayak in their findings. Notably, similar spatial expression patterns have been demonstrated for Alx using FISH in Pan et al., available on BioRxiv. Given the importance of Kayak in the presented arguments, it is advisable to also investigate its spatial expression patterns using techniques such as FISH.

    Significance

    The paper introduces novelties to the field of regeneration and developmental biology by leveraging Craspedacusta polyp as a novel model system for investigating the evolutionary and developmental dynamics of tentacles. In doing so, it sheds new light on the intricate mechanisms underlying tentacle formation and patterning. Furthermore, the study implicates Kayak in the regulation of Wnt3, adding a fresh perspective to our understanding of the molecular pathways governing Hydra regeneration. Notably, the results of the research challenge the prevailing notion of autoregulation of Wnt3, which has long been considered fundamental to organizer formation in Hydra. While these findings offer intriguing insights, further investigation will be crucial to conclusively ascertain the validity of this assertion.

    Despite the clarity of the data presented, the interpretation and integration of these findings in the manuscript are lacking. The narrative at times feels disjointed, with different storylines loosely connected. While the findings are intriguing and merit publication, a substantial revision of the manuscript is necessary to provide a more coherent and illuminating interpretation of the results.

    The implications of this research extend beyond the specific confines of Craspedacusta polyp and Hydra biology. It holds significant relevance for both the Hydra biology community and the broader field of Notch signaling research. By highlighting the pivotal role of Notch signaling in regeneration and patterning within Hydra, the study enriches our comprehension of this model organism and its evolutionary adaptations. Moreover, it provides a valuable lens through which the evolution of Notch signaling cascades can be examined. This interdisciplinary approach underscores the interconnectedness of diverse biological systems and underscores the importance of exploring novel model organisms to unravel the complexities of evolution and development.

  5. 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

    The study investigates the role of Notch and beta-catenin signaling in coordinating head regeneration in Hydra. It combines gene expression dynamics, inhibitor treatments, and comparisons with Craspedacusta polyps to propose a lateral inhibition model for Notch function during Hydra head regeneration, mediating between two pattern-forming systems.

    Three main concerns arise from this work:

    1. Lack of spatial expression data: The study proposes a model based on pattern-forming systems but falls short of providing direct spatial expression data for the genes under consideration in both control and treated scenarios. This gap weakens the empirical support for the proposed model.
    2. Clarity and relevance of Craspedacusta comparisons: The section discussing the regeneration in Craspedacusta polyps appears somewhat disjointed from the main narrative, with its contribution to the overarching story of Hydra regeneration remaining unclear.
    3. Accessibility of the text: The study's presentation, including its title, abstract, and main text, presents challenges in terms of clarity and accessibility, making it difficult for readers to follow and understand the research's scope, methodologies, and conclusions.

    Significance

    In conclusion, while the study aims to advance our understanding of the complex signaling pathways governing Hydra head regeneration, it necessitates significant revisions. Enhancing the empirical evidence through detailed spatial patterning data, clarifying the comparative analysis with Craspedacusta polyps, and refining the narrative to improve accessibility are critical steps needed to solidify the study's contributions to the field.

  6. 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

    Steichele et al tackle a long standing question about the precise role of Notch signalling during Hydra head regeneration. They compare the inhibition of Wnt signalling and Notch signalling by pharmacological inhibition. The authors show that inhibition of Wnt signalling blocks tentacle formation but not formation of the hypostome, wnt3 expression or organizer activity. The authors further attempt to understand this in a comparative sense using Craspedacusta. In addition, the authors propose HyKayak as a potential repressor of Wnt3 expression downstream of Notch signalling. there are howerver a number of major and minor problems with the study which must be addressed before it is suitable for publication as outlined below:

    Major:

    • The introduction is lacking a full description of what is known about transcriptional changes during Hydra regeneration and in particular the role of Wnt signalling in this process. Of note the authors do not cite several important studies from recent years including (but not limited to):

    https://doi.org/10.1073/pnas.2204122119

    https://doi.org/10.1186/s13072-020-00364-6

    https://doi.org/10.1101/587147

    https://doi.org/10.7554/eLife.60562

    This problem is further compounded later when the authors do not cite/discuss work which has performed the same or similar analyses to their own. The authors should endeavor to give a more comprehensive background knowledge.

    • The authors do not cite or reference at all the study by Cazet et al. which used iCRT14 along with RNAseq and ATACseq to probe the role of Wnt signaling during early regeneration. This is a major issue. Although I appreciate that the authors have done much longer time courses and that their data therefore add something to our understanding it will still be important to discuss here. For example, the authors show that Wnt3 is activated normally in iCRT14 animals. Is this also seen in the RNAseq from Cazet et al.,
    • The visualizations used in Figure 3 are quite difficult to interpret and do to in all cases match descriptions in the text. The way the same type data is displayed in figure 5 si much nices.It is also better to treat the same types of data in the same manner consistently throught the paper. For Hes, for example, the authors state that there is a reduction although the data shows that this is very small and taking into account the 95% confidence interval does not seem to be significant. If this is the case then the positive control is not working in this experiment. This would be much clear if individual time points were compared like in figure 5 and statistical tests shown. The authors then state that Alx is not affected but there is actually a larger effect than what they deemed significant for Hes ( the axes are notably different between these two and I think a more consistent axis would make the genes more comparable). Similarly, Gsc is described as being not affected at 8 hours but it appears again to change more that the positive control Hes. Given this I would call into question the validity of this dataset and/or the interpretation by the authors. A more thorough analysis including taking better into account statistical significance would go along way to increasing confidence in this data.
    • The same issues in interpretation described for Figure 3 also apply to figure 4. The authors state that Wnt7 is affected less than Wnt1 and 3 but this is not evident in the figure and no comparative analysis is performed to confirm this. The same for Wnt 11 and 9/10c where what the authors description is very difficult to see in the figure. Spt5 is apparently upregulated, but this is not discussed. Again the axes are ntably different making it even more difficult to compare between samples.
    • In their description of figure 4 the authors completely omit to discuss the Cazet et al dataset which has the exact same early timepoints for iCRT14 treatment. This must be discussed and compared and any difference noted.
    • End of page 11: The authors propose a model thereby the role of Notch in Wnt3 expression may be due to the presence of a repressor. However, I don't see any putative evidence at that stage. The authors also do not cite relevant work from both Cazet et al. and Tursch et al which show that Wnt3 is likely upregulated by bZIP TFs. In both these cases the authors show evidence of bZIP TF binding sites in the Wnt3 promoter along with other analyses. This is very relevant to the model presented by the authors here and must be discussed and compared. In particular the authors put forward HyKayak as an inhibitor of Wnt3 and this should be discussed along with the previous work.
    • On page 12 the authors conclude based on gene expression in inhibitor treatment that there is a change in complex composition of the two transcription factors. This is something which would require biochemical evidence and I therefore suggest they remove this entirely.
    • The authors use experiments in Craspedacusta to test their hypothesis of the role of Wnt and Notch signaling in Hydra. There is, in my opinion, an incorrect logic here. Regardless of the outcome, the roles of Wnt and Notch could conceivably be different in the two species and therefore testing hypothesis from one is not possible in the other. The authors should reframe their discussion of this to be more of a comparative framework. Moreover, the results do not necessarily indicate what the authors say. In Hydra Notch signaling is required to form the hypostome/mouth and this is not the case in Craspedacusta while Wnt signaling is required. The authors do not cite an important study from another Hydrozoan Hydractinia (Gahan at al.,2017). In that study the authors show that DAPT inhibits tentacles during regeneration but that the hypostome (or at least the arrangement of neurons and cnidocytes around the mouth) forms normally. This would indicate that in Hydractinia the process of head formation is different to in Hydra and would be congruent with what is shown here in Craspedacusta. This should be more thoroughly discussed, and all relevant literature cited.
    • From reading the manuscript I do not fully understand the model the authors put forward. It is unclear what "coordinating two independent pattern forming systems" really means. It might be beneficial to make a schematic illustration of the model and how it goes wrong in both sets of inhibitor treatments.

    Minor:

    • The abstract could be rewritten to have more of an introduction to the problem rather than jumping directly into results. It would also be beneficial if the abstract followed the logic of the paper.
    • In Figure 3 and 4 it is not clear why they are divided into A and B. It appears that the categorization of genes into different groups lacks a clear rationale .This seems totally unnecessary. In addition, the order in which the genes are described in the text does not match what is seen in the figure making it confusing to follow.
    • In Figure 5 the authors use two different types of charts and I would stick with one. B is much better as it shows the individual data points as well as other information. I would use this throughout including in Figure 3 and 4.
    • Figure S3 is missing a description of panel C.
    • In figure S3 it is not clear why the inhibitor was removed and not kept on throughout the experiment. Please discuss.
    • Figure S4 has no A or B in the figure, only in the legend.

    Significance

    Although some of the authors data appear to be novel I find the study makes only minor progress on the questions. In particular the authors do not properly cite the relevant literature and to put their manuscript into the correct context. The new model proposed by the authors is based entirely on qPCR data which is not thoroughly analyzed and are not strong enough in the absence of information about the spatial expression the genes they discuss. The proposal of HyKayak as a negative regulator of Wnt3 is interesting but the authors do not provide any solid direct evidence for this (ChIP, EMSA etc) and it is somewhat in disagreement with other models of bZIP function in the literature (which again are not discussed).

    The manuscript is of limited general interest. It has a number of interesting observations which would be of interest to the Hydra community and the broader cnidarian community. The study lacks contextualization within a broader framework, whether it be in the context of regeneration or Wnt/Notch signaling. This limitation may narrow the overall interest in it.

  7. Version published to 10.1101/2024.02.02.578611v1 on bioRxiv