Mutations were generated in the HS3545 background, which ubiquitously expresses the TIR1(F79G) receptor,
love the power of auxin/aid system in this paper! If you ever want/need cell-type specific control also, we made a bunch of tools in https://doi.org/10.1093/genetics/iyad013 that might be super useful so you could visualize cell cycle state and get robust auxin-mediated depletion in whatever lineage you want (would just need to pair this with seam-cell promoter FLP).
In control populations under these conditions, up to 10% of the population will survive this heat shock. However, after CDK-4 ablation 56% of the population survived the same heat shock treatment (Figure 5B). Although LIN-35 ablation had no significant effect on heat stress survival, co-ablation with CDK-4 suppressed survival to wildtype levels (Figure 5B). These results suggest that when nucleoli are large and protein synthesis is high, the animal is better equipped to survive phases of acute heat stress.
This is such an amazing result! It makes me wonder if tuning CDK-4 activity (or maybe just nucleolar size in general???) could be an evolutionary selection criteria for thermotolerance in higher temperature environments? Are there wild isolates of C. elegans from warm climates that have bigger nucleoli than cold-isolated isolates (or just N2?). Or do you think the negative selection on aging might counter-balance this?
Yet, while wildtype polyps start to spawn after about 4-5 months postfertilization, even after 1 year none of the mutants spawned any gametes, neither oocytes nor sperm, suggesting that the nanos2 mutants are sterile.
Have you looked to see if your nanos2 mutants regenerate? Seems like the perfect genetic background to dissect the requirements for regeneration in animals that have (artificially) lost the ability to reproduce sexually!
In the single cell spreads we can again detect nanos2::mOrange expression in all neuroglandular cell types (Fig. 2H-M). However, in both nanos2::mOrange and piwi1::mOrange transgenic animals, only a small fraction of neuroglandular cell types show low levels of residual fluorophore, and most are not expressing mOrange. Therefore, we conclude that both genes are most likely expressed in the precursor population, from which the neuroglandular cell types differentiate and carry on residual mOrange reporter protein.
This is really interesting! From your methods it looks like you have to fix the single cell spreads for imaging - if it was possible to do this imaging live you could try to test this hypothesis explicitly by photobleaching/FRAP and see if any new mOrange protein is detectable in these mOrange+ cells.
to continue to produce LAG-2 protein
in future work it would be really cool to use the auxin system to have a means of rapidly degrading LAG-2 protein in both niches (assuming that TIR1 gets inherited into the patch of. course) - but if it does, it would give you a way of functionally testing the temporal aspect - if you degrade the LAG-2 early vs late...
coexpressing
if it's helpful to point out, the lag-2 bmb202 allele is an endogenous read-out of lag-2 transcription since it's knocked in at the lag-2 locus.
Figure 2
i love this figure! Since your RNAi and mutants cause a range of phenotypes, it would be useful to put the penetrance for the representative image in the corner of the figure (e.g., x/y, %). You could also do this in any other figure with images showing a phenotype.
a distal tip cell membrane marker
just to help with clarity, mention that it's an mNG marker here, so when you bring up mNG below the connection is immediately obvious
Therefore, wealso tried to block the ectopic expression of germline genes by performing RNAinterference against the germline transcription factor LSL-1
if you do future studies with this (I hope so, it's so cool!), you could generate a degron version of LSL-1 to potentially have more acute control - you could also then figure out site of action by pairing it with different tissue/cell type specific FLPs (if you used the FLP/FRT TIR1 system).
One possibility is that some normal component ofneuronal function is blocked at the transcription level.
this result is so fascinating! for future work it would be really cool to nail the site of action for the behavioral change - I wonder if you could generate a reporter line or a knock-in of a germline gene that you think is being mis-expressed paired with GCaMP so you could look for neurons that are mis-expressing germline genes and a loss of signal transduction/activity?
spiralian development
so awesome to see functional work happening in this amazing clade of organisms! I'm really amazed that you see such broad effects from the oe dnBMPR1 construct across germ layers! I'm super curious if there are lineage/germ-layer specific mechanisms that are deployed that normally facilitate correct BMP signaling associated with fate/patterning?
To determine if BMPR1ΔK was expressed in embryos
Do you know if wild-type endogenous BMPR1 is expressed ubiquitously? Would you expect it to be? What happens in a cell that isn't expressing endogenous BMPR1 if it sees high levels of the dnBMPR1? I guess the answer would be nothing if the downstream components for BMP signaling aren't deployed in that cell either?
In some uninjected controls, nuclear pSMAD1/5/8 levels varied between non-dividing cells; some nuclei appeared to have higher levels of pSMAD1/5/8 compared to their neighbors (cyan versus white arrows in Fig. 2).
this is interesting! Was this in any sort of lineage specific pattern?
fgt-1
just in case it's at all useful, there is an mNG knockin available of fgt-1 at the CGC here: https://cgc.umn.edu/strain/NK2540
suggesting there there may also occur a wider, albeit incomplete, dauer rescue in some tissues.
First, I just want to say that this is a really amazing body of work and I really enjoyed reading your pre-print!
I've thought a lot about quiescence / cell cycle entry / cell fate in the L3 reproductive system, and there might be some useful tools that could get you a tissue-specific, sensitive read-out of cell cycle entry. My former lab made some cell cycle sensors that change nuclear-->cytosolic localization from G0-->G1-->S-->G2 (this strain: https://cgc.umn.edu/strain/DQM543 from https://elifesciences.org/articles/63265) would be a sensitive marker of the change from G0 in the P cells to G1 if you had the right microscope.
It might be really interesting to look at this CDK activity sensor in situations where some tissues are in dauer and some are not!
The first sign of acceleration of the sequence of morphological transformations of the anchor cell on MCPs was evident around 36 hours
see my comment on the staging in the methods - i think it's important to stage AC invasion by multiple criteria - DIC showing the VPC stage is an excellent one in addition to any morphological changes you might see in the AC, which can be highly variable during invasion.
Staging (see Figure S1D for representative images) was based primarily on the shape of the cell’s ventral side. AC1 = long curving ventral side not attached to P6 daughter cells. AC2 = flat and short ventral side indicates attachment. AC3 = invasive protrusion forms a V-shape. AC4 = invasive protrusion retracts forming an M-shape.
There is a well-established staging system for the timing of AC invasion based on the divisions of the underlying P6.p lineage (see Sherwood et al. 2003 or later papers from Sherwood lab). The shape of the AC is indicative but also highly variable. It is also worth pointing out that the AC breaches the underlying basement membrane during the mid-L3 stage, and that in wild-type the AC is in contact with the underlying vulval cells prior to the L3/L4 molt. The reason this is important, is that it appears from your images that your AC4 image is younger (by VPC morphology) than your AC3 image (although I realize that the DIC image is quite small in the figure so I'm not 100% sure). Given that the effect you are seeing from the MCPs in relation to AC morphology (as a proxy for invasion status) is getting to "AC4" faster, I think it would be important to be careful with developmental staging here. I'd be happy to look at your images if that would be at all helpful!
vulva anchor cell,
technically, the anchor cell is part of the somatic gonad, it just induces vulval fates (and forms the connection between the uterus and vulva)
We propose that the linker cell chromosomes help anchor the cytoskeleton and membrane adhesion molecules, making the nucleus a huge anchor that can withstand the force of collective cell migration
This would be an interesting model to test for sure! Have you thought about whether the SMC complex is functioning in the LC to regulate the quiescent state? For example in yeast in appears to play a critical role in quiescent cells to regulate chromatin state and is important for transcriptional repression: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6368455/
it would be interesting to test this model in the LC! Of course it could be doing both - playing some sort of functional tethering/physical role as well as maintaining genome integrity to not activate genes inappropriately.
Images are of animals in the L4 larvae stage, with lag-2p::YFP (syIs128) (colored in green) labeling the detached linker cell (LC).
It might be nice to add in an image of wild-type showing the LC attached to the trailing male gonad here to compare to the detached phenotypes of the mutant alleles
G2 arrest
First, I just want to say that I love this preprint! I've been thinking about interphase cell cycle checkpoints associated with either cell fate or morphogenesis for awhile now. I can think of two other contexts where cells arrest in G2 during development that might be relevant for you to think about in relation to this study:
neural stem cells in Drosophila are arrested in G2 (from the Brand lab (see doi: 10.1126/science.aan8795 and doi.org/10.1016/j.devcel.2019.02.015)
During convergent extension in Xenopus, G2 arrest is critical for proper morphogenesis (see 10.1242/dev.01054).
interphase
Have you thought about how different interphase states (G1/S/G2) might affect the underlying biology? For example, it could be that cells localize proteins in unique ways in G1 or G2 and that my taking your measurements en masse, you might miss out on the subtleties.
EMT was reported to be connected to enhanced cell migration and cell proliferation in metastatic cancer cells
there is a really interesting literature "Go vs. Grow" that suggests that when cells adopt invasive/migratory phenotypes, such as undergoing EMT, that they arrest in the cell cycle. This creates a dichotomy between proliferation and invasion, suggesting that from a cell biological perspective, these cell behaviors may be mutually exclusive, at least in some contexts.
we worked with rounded, non-adherent cells since this has the advantage that cell shapes are spherical in both epithelial and EMT-transformed conditions with a largely uniform actin cortex. In this way, a meaningful comparative analysis of cortical protein association between epithelial and the mesenchymal-like cells becomes possible.
I appreciate trying to generate equivalent conditions to do your comparison, but I am confused as to how rounded cells are equivalent to cells that have adopted a mesenchymal migratory/invasive morphology? Specifically because cells undergoing EMT need to undergo massive polarization changes from their epithelial state.
G2 phase
I wonder if the G2 arrest is a functional requirement for re-attachment? It seems like it is likely, but has anyone ever mis-expressed a cyclin dependent kinase inhibitor (dacapo?) in the GSCs to lock them into G1/G0 and assess their ability to dock? I'm not sure how technically feasible this experiment is - it also might get at my earlier question as to whether or not the returning GBs need to be in G2 to migrate back?
dedifferentiating gonialblast
from your movie it looks like the GB is also migrating in G2? Is this always the case? That they might be undergoing a prolonged G2 arrest - that might be important for their migration too? There is some really interesting examples in the literature of cells that need to be in certain cell cycle states to execute specific morphogenetic behaviors - it would be interesting if this migration is one of these too!
Note that G1 phase was not detectable.
Is this due to the G1 phase being too short for the time interval for imaging?
red)
consider showing this figure in green:magenta to make it easier to interpret for color blind individuals
Future work will explore how dysregulated pHi dynamics in the mother cell alters or modulates daughter cell outcomes.
This would be really cool to look at, and was something I kept thinking about as I was reading the results section - what happens to pHi at mitotic exit in cells with asymmetric fates - i.e., if one daughter exits into a CDK-low state and the other exits into a CDK-inc state? Have you thought about just pairing your pHi biosensor with the ratiometric CDK sensor, instead of FUCCI, so you could simultaneously monitor cell cycle state quantitatively, pHi ratios, and distinguish quiescent cells shortly after mitotic exit from G1 cells that will likely cycle again?
From these data, we conclude that pHi is dynamic through the cell cycle at the single-cell level: pHi decreases during G1/S, increases in early S phase, decreases leading to S/G2, increases prior to G2/M, and decreases following mitosis.
I know you have a summary figure at the end, but it would be helpful to just include this as a diagram/schematic at the end of Fig. 3 if there's room.
Representative images of pHi measurements
Have you considered displaying your heat maps with a color-blind friendly LUT - I think the data shown in the middle column would be very difficult to parse if you were red-green color blind, but is really quite striking!
aggregates
I was wondering if the use of mCherry led to aggregates - has there been any thought of redesigning the biosensor with a monomer like mKate2 or mScarlet-I (if you need to keep it in the red/near-far-red range) so that you don't have to worry about this aspect of localization issues?
Cell proliferation requires FGF signaling, whereas the three-dimensional orientation of the organ depends on Wnt signaling. Specification and maintenance of tube cell fate requires Delta/Notch signaling.
Consider putting these in the order you reported them in the results (like my suggestion for the highlights below)
❖ The Wnt pathway controls directional outgrowth❖ The FGF pathway promotes regionalized cell proliferation❖ The Notch/Delta pathway is essential in cell fate repression in tubulogenesis
I would consider re-ordering these to mirror the order you present them in the results - Notch/Delta --> Wnt --> FGF --> FGF targets
Although the basal lamina was generally contiguous with the tube epithelium, we observed a laminin protrusion on the anterior side of the tube that was devoid of cells (Figure 1 E, F arrows), suggesting that cells left the laminin network on the anterior side after their migration toward the posterior end.
I am so fascinated by this - did you see cells help make this protrusion and then leave? What is doing this if the region is acellular? And is there a L/R bias like it looks like in Fig. 1E?
We found that in Wnt-inhibited larvae, the tubes elongated towards the dorsal ectoderm instead of posteriorly, as in controls
Any clue why in the absence of Wnt the tube still has an orientation (dorsal) but just to the wrong place rather than just a random orientation? Is there some signal that in the absence of Wnt that draws the tube dorsal (Nodal/BMP or something like this?)
To define the nature of sea star tube formation, we took advantage of the optical clarity of the larvae and developed techniques to immobilize and live image morphogenesis continuously from gastrulation to early larval stages (Figure 1- Video 1)
this is SO amazing! Have you considered writing up a protocol for protocols.io or something similar to highlight how cool this is!
Since the hydropore canal forms by tissue bending at a defined branch point, we propose that Six1/2 activation downstream of FGF signaling is critical for branching morphogenesis.
woah! I wonder if it's sufficient to induce extra branching morphogenesis if you OE it spatially/temporally??? (not for this paper - there's so much amazing biology in this paper… do this in your own lab! :). )
These results suggest that continued Delta/Notch signaling is important to restrict excess muscle specification.
this is really really interesting - and would be fascinating to follow up on as a model for transdifferentiation!
In addition, exploring the limitations of knock-in fragment size would be useful to determine if the approach could be used for multi-gene/pathway knock-in at neutral genomic sites for biotechnology applications.
one idea - you could explore split FPs - you could express GFP(1-10) off the episome and then just KI in GFP11's (there are several split FPs now available) - this could be a rapid way to label multiple things (though you would need to maintain the line using NAT selection) - which brings me to my last question - have you thought about making a landing pad for site specific insertion now that you can insert into defined locations? There would be all sorts of great use cases for this!
It has been demonstrated in other organisms that HDR can be offset from the cut-site which would then enable GFP to be inserted directly prior to the stop codon and increase sgRNA flexibility. However, knock-in efficiency can rapidly drop off as the cut-site to HDR-site distance increases39.
Yes! See my comment above - I definitely think, given the kind of efficiency you report here, that you can explore moving your insertion site to the N- or C-terminus of your target gene
Using CRISPick28 (see methods) we selected a sgRNA with a predicted on-target score of 0.7 that cut 40 bp upstream of the BST2 stop codon. We designed homology arms that covered both sides of the predicted Cas9 cut site (Fig. 2a)
Based on lots of tagging of endogenous loci using a similar plasmid based delivery system in C. elegans, I would think that you should be able to get away with moving the insertion site to the C-terminus (in this case) or N-terminus (in other cases) - it would be great to test this out - but we (and others) have found that if you re-code the homology arm in your repair plasmid between the cut site and insertion point, you can get KIs when you push the distance between the cut site and the insertion site, avoiding having to insert your FP into the gene. Is there a locus of interest with some different sgRNAs at the C- or N-terminus that you could experimentally try this out with (like one that would split the insertion site, one that would be ~10-20 bp away, and one that would be ~50-60bp away)?
We developed a novel reporter stable cell line using the Fucci reporter system for visualization of the cell cycle (Fig. 6A and Fig. S6A). Due to the limitation of available fluorescence channels, we used only the Cdt1 (30-120) fragment, which is stabilized in G1 phase and degraded in S/G2/M phase (Sakaue-Sawano et al., 2008).
One idea that would allow for greater resolving power of cell cycle kinetics while simultaneously monitoring p53 transcription would be to engineer the DNA Helicase B based (DHB) CDK1/2 biosensor (Spencer et al. 2013 doi: 10.1016/j.cell.2013.08.062) into the last position fused to mCerulean instead of the CDT1 component of the FUCCI system. The added benefit, after bookmarking the CDK sensor, would be a more precise read-out of cell cycle state using DHB as well as discriminating between cells that exit in G0 (cdk-low) vs G1 (cdk-increasing). If you wanted an even more precise construct for delineating cell cycle state, you could consider swapping the NLS-mCherry for PCNA::mScarlet-I, as it seems like there is a lot of interesting biology associated with S-phase in your p53 sensor in this pre-print, and you could precisely define S-phase based on nuclear puncta formation in PCNA (I might avoid mCherry in this case due to mCherry aggregates) while still maintaining the nuclear mask, as PCNA is expressed and nuclear localized throughout the cell cycle.
This suggests that from the perspective of ectoderm patterning, echinoderms are mostly head-like animals, and prompts a reinterpretation of the evolutionary trends that made echinoderms the most derived animal group.
This is a really exciting idea - I was hoping to read more about why you felt like this was the key take-away based on the data of the patterning system of the ambulacral ectoderm in the discussion section? Also, this is reminiscent of tardigrades as head-like animals, missing the Hox trunk genes from Bob Goldstein's lab and I think some of Andi Hejnol's work on many spiralian taxa - it feels like something important evolutionary is going on here with a many metazoan taxa utilizing the deuterostome/vertebrate "head" patterning system, though perhaps we should be thinking about the deuterostome/vertebrate co-option of this system???
These marker genes included transcription factors
Did you find any new markers with regional/cell-type specificity from your spatial transcriptomics (maybe this is the basis for future work and if so, great!)
b,
minor comment, here and elsewhere, don't forget your scale bars!
The uncoupling of an ectodermal head and trunk programs is not unique to P. miniata and has been demonstrated in both larval echinoderms and hemichordates39,47, and recently in annelid larvae43, suggesting that these regulatory programs can be uncoupled over macroevolutionary time frames.
from my comment in the abstract - this seems like the perfect place to talk a little more about the "adult sea stars are heads" hypothesis
The accumulation pattern of TNY1 protein throughout the cell cycle was determined in a wild- type culture synchronized under the standard 12hr:12hr light:dark diurnal cycle.
Would it have been possible to look at this orthogonally by visualizing and quantifying the localization pattern of the TNY1-mCherry fusion protein?
Figure 3.
have you thought about including a summary schematic at the end of figure 3 (you have space for it) - i like the schematic in A that sets up the figure, it would be nice to have the results summarized at the end.
Video S3)
It would be helpful to add some annotation to this supplemental video - time stamp, arrows, etc.
These marker genes included transcription factors
Did you find any new markers with regional/cell-type specificity from your spatial transcriptomics (maybe this is the basis for future work and if so, great!)
The uncoupling of an ectodermal head and trunk programs is not unique to P. miniata and has been demonstrated in both larval echinoderms and hemichordates39,47, and recently in annelid larvae43, suggesting that these regulatory programs can be uncoupled over macroevolutionary time frames.
from my comment in the abstract - this seems like the perfect place to talk a little more about the "adult sea stars are heads" hypothesis
b,
minor comment, here and elsewhere, don't forget your scale bars!
This suggests that from the perspective of ectoderm patterning, echinoderms are mostly head-like animals, and prompts a reinterpretation of the evolutionary trends that made echinoderms the most derived animal group.
This is a really exciting idea - I was hoping to read more about why you felt like this was the key take-away based on the data of the patterning system of the ambulacral ectoderm in the discussion section? Also, this is reminiscent of tardigrades as head-like animals, missing the Hox trunk genes from Bob Goldstein's lab and I think some of Andi Hejnol's work on many spiralian taxa - it feels like something important evolutionary is going on here with a many metazoan taxa utilizing the deuterostome/vertebrate "head" patterning system, though perhaps we should be thinking about the deuterostome/vertebrate co-option of this system???
Figure 3.
have you thought about including a summary schematic at the end of figure 3 (you have space for it) - i like the schematic in A that sets up the figure, it would be nice to have the results summarized at the end.
Video S3)
It would be helpful to add some annotation to this supplemental video - time stamp, arrows, etc.
The accumulation pattern of TNY1 protein throughout the cell cycle was determined in a wild- type culture synchronized under the standard 12hr:12hr light:dark diurnal cycle.
Would it have been possible to look at this orthogonally by visualizing and quantifying the localization pattern of the TNY1-mCherry fusion protein?
We developed a novel reporter stable cell line using the Fucci reporter system for visualization of the cell cycle (Fig. 6A and Fig. S6A). Due to the limitation of available fluorescence channels, we used only the Cdt1 (30-120) fragment, which is stabilized in G1 phase and degraded in S/G2/M phase (Sakaue-Sawano et al., 2008).
One idea that would allow for greater resolving power of cell cycle kinetics while simultaneously monitoring p53 transcription would be to engineer the DNA Helicase B based (DHB) CDK1/2 biosensor (Spencer et al. 2013 doi: 10.1016/j.cell.2013.08.062) into the last position fused to mCerulean instead of the CDT1 component of the FUCCI system. The added benefit, after bookmarking the CDK sensor, would be a more precise read-out of cell cycle state using DHB as well as discriminating between cells that exit in G0 (cdk-low) vs G1 (cdk-increasing). If you wanted an even more precise construct for delineating cell cycle state, you could consider swapping the NLS-mCherry for PCNA::mScarlet-I, as it seems like there is a lot of interesting biology associated with S-phase in your p53 sensor in this pre-print, and you could precisely define S-phase based on nuclear puncta formation in PCNA (I might avoid mCherry in this case due to mCherry aggregates) while still maintaining the nuclear mask, as PCNA is expressed and nuclear localized throughout the cell cycle.
It has been demonstrated in other organisms that HDR can be offset from the cut-site which would then enable GFP to be inserted directly prior to the stop codon and increase sgRNA flexibility. However, knock-in efficiency can rapidly drop off as the cut-site to HDR-site distance increases39.
Yes! See my comment above - I definitely think, given the kind of efficiency you report here, that you can explore moving your insertion site to the N- or C-terminus of your target gene
In addition, exploring the limitations of knock-in fragment size would be useful to determine if the approach could be used for multi-gene/pathway knock-in at neutral genomic sites for biotechnology applications.
one idea - you could explore split FPs - you could express GFP(1-10) off the episome and then just KI in GFP11's (there are several split FPs now available) - this could be a rapid way to label multiple things (though you would need to maintain the line using NAT selection) - which brings me to my last question - have you thought about making a landing pad for site specific insertion now that you can insert into defined locations? There would be all sorts of great use cases for this!
Using CRISPick28 (see methods) we selected a sgRNA with a predicted on-target score of 0.7 that cut 40 bp upstream of the BST2 stop codon. We designed homology arms that covered both sides of the predicted Cas9 cut site (Fig. 2a)
Based on lots of tagging of endogenous loci using a similar plasmid based delivery system in C. elegans, I would think that you should be able to get away with moving the insertion site to the C-terminus (in this case) or N-terminus (in other cases) - it would be great to test this out - but we (and others) have found that if you re-code the homology arm in your repair plasmid between the cut site and insertion point, you can get KIs when you push the distance between the cut site and the insertion site, avoiding having to insert your FP into the gene. Is there a locus of interest with some different sgRNAs at the C- or N-terminus that you could experimentally try this out with (like one that would split the insertion site, one that would be ~10-20 bp away, and one that would be ~50-60bp away)?
Since the hydropore canal forms by tissue bending at a defined branch point, we propose that Six1/2 activation downstream of FGF signaling is critical for branching morphogenesis.
woah! I wonder if it's sufficient to induce extra branching morphogenesis if you OE it spatially/temporally??? (not for this paper - there's so much amazing biology in this paper… do this in your own lab! :). )
We found that in Wnt-inhibited larvae, the tubes elongated towards the dorsal ectoderm instead of posteriorly, as in controls
Any clue why in the absence of Wnt the tube still has an orientation (dorsal) but just to the wrong place rather than just a random orientation? Is there some signal that in the absence of Wnt that draws the tube dorsal (Nodal/BMP or something like this?)
These results suggest that continued Delta/Notch signaling is important to restrict excess muscle specification.
this is really really interesting - and would be fascinating to follow up on as a model for transdifferentiation!
Although the basal lamina was generally contiguous with the tube epithelium, we observed a laminin protrusion on the anterior side of the tube that was devoid of cells (Figure 1 E, F arrows), suggesting that cells left the laminin network on the anterior side after their migration toward the posterior end.
I am so fascinated by this - did you see cells help make this protrusion and then leave? What is doing this if the region is acellular? And is there a L/R bias like it looks like in Fig. 1E?
To define the nature of sea star tube formation, we took advantage of the optical clarity of the larvae and developed techniques to immobilize and live image morphogenesis continuously from gastrulation to early larval stages (Figure 1- Video 1)
this is SO amazing! Have you considered writing up a protocol for protocols.io or something similar to highlight how cool this is!
❖ The Wnt pathway controls directional outgrowth❖ The FGF pathway promotes regionalized cell proliferation❖ The Notch/Delta pathway is essential in cell fate repression in tubulogenesis
I would consider re-ordering these to mirror the order you present them in the results - Notch/Delta --> Wnt --> FGF --> FGF targets
Cell proliferation requires FGF signaling, whereas the three-dimensional orientation of the organ depends on Wnt signaling. Specification and maintenance of tube cell fate requires Delta/Notch signaling.
Consider putting these in the order you reported them in the results (like my suggestion for the highlights below)