Germ layer-specific regulation of cell polarity and adhesion gives insight into the evolution of mesoderm

[…] Given the diversity of gastrulation mechanisms in Cnidaria, examining whether other cnidarians utilize EMT during gastrulation should be informative. As EMT behaviors have been described descriptively during regeneration and trans-differentiation in sponges (see Coutinho et al., 2017) and Nakanishi et al., 2014), it is plausible that other Cnidaria may utilize EMT during gastrulation, or that at EMT may have evolved prior to the Cnidarian-Bilaterian divergence instead. The implications of this work are not only relevant to our understanding of the evolution of mesoderm, but to cell biologists interested in the evolution and plasticity of EMT, which is a critical cell behavior during development and disease states such as cancer dissemination.

We completely agree with this comment and further address this issue in the Conclusion section.

Essential revisions:

1) in subsection “The NvaPKC/Par complex regulates the formation and maintenance of cell-cell junctions” the authors write: "dnNvPar-1 can be phosphorylated by aPKC but cannot phosphorylate the aPKC/Par complex (Böhm, Brinkmann, Drab, Henske, and Kurzchalia, 1997; Vaccari, Rabouille, and Ephrussi, 2005). Thus, dnNvPar-1 can localize to the cell cortex where aPKC is inactive." However, the authors do not have any evidence for phosphorylation/kinase activity for Nematostella dnPar-1. The authors should qualify this statement or provide experimental evidence.

The reviewers are correct. We have added an additional figure where we show co-IP experiments using the NvPar-1 specific antibody (Figure 3—figure supplement 2) showing that NvPar-1 interacts with NvaPKC and NvPar-6 (both proteins detected with their own specific antibodies). In addition, we observed three bands labeled with NvPar-1 antibody around 80 KD, suggesting different phosphorylation states of this protein, which may be a product of NvaPKC activity.

However, we were not able to perform co-IP experiments on the dominant negative form of NvPar-1 because of technical difficulties. To perform the co-IP, we needed to extract protein from over 13,000 wild type embryos; we were not able to inject and express the dominant negative form of this protein into this large number of embryos. Hence, we changed our statement to:(subsection “The NvaPKC/Par complex regulates the formation and maintenance of cell-cell junctions”) “Since NvPar-1 is phosphorylated by NvaPKC (Figure 3—figure supplement 2), we predict that, as in other systems, dnNvPar-1 could be phosphorylated by NvaPKC but would not phosphorylate the NvaPKC/Par complex [37,38]. Thus, dnNvPar-1 can localize to the cell cortex where aPKC may be inactive.”

2) In subsection “The NvaPKC/Par complex regulates the formation and maintenance of cell-cell junctions”, the authors conclude that Par3 KO has no significant effect on lineage markers; however, Figure 3—figure supplement 1E is not quite consistent with this notion, in particular since the percentages of KO embryos shown on the PCR gel on Figure 2—figure supplement 1 is low (2/12); it is of course possible that animals with a wild type size band (i.e., about 985 bp) are also mutated (e.g. small deletion resulting in a frame shift), but this isn't known since the authors didn't sequence the PCR products. Hence, the embryos with defective lineage marker expression could be the mutated ones.

As we show, the endomesoderm and the overall morphology of the embryos were drastically affected by Par3 KO making it hard to interpret the phenotype. The complete zygotic deletion of the proteins NvPar6 and NvPar3 resulted in lethal phenotypes (embryos developed with maternal contributions during earlier stages) and since all stable mutant embryos could have had some level of mosaicism where the wild type gene/protein may have been present in their ectoderm, we chose to assess our experiments by ISH and IHC using our specific probes, a specific antibody against NvPar-6, and a ß-catenin antibody.

In order to make this clearer, we have rephrased this statement:

(subsection “The NvaPKC/Par complex regulates the formation and maintenance of cell-cell junctions”) “Although it was difficult to dissect significant changes in the expression of germ layer markers (e.g. Nvbra, Nvsnail, NvSix3/6, and Nvfz10) from the morphological changes associated with epithelial integrity when these genes were disrupted (Figure 3—figure supplement 4E), it is clear that the primary defect in NvPar3 KO were aspects of cell adhesion and not cell type specification.”

3) Figure 3—figure supplement2A – The change in epithelial integrity of the ectoderm following Par6 and Par3 disruption is quite stunning. It would be important for the authors to quantify the range of thickness by measuring as compared to control embryos and then use the appropriate statistical test to show that this is a significant result. A dot plot may be the most effective way to display the range of phenotypes observed in thickness of the epithelium. In other words, measuring thickness of the epithelium in a statistical framework. will benefit the work greatly.

Done, Figure 3—figure supplement 3.

4) Figure 3—figure supplement 1; what do the thicker and thinner than normal epithelia reflect? Is it just being abnormal, or is there a different interpretation for thick vs thin?

With the information that we have, we can only say that is an abnormality caused by the loss of epithelial homeostasis. We have added a note to the figure legend in Figure 3—figure supplement 3.

5) In Figure 5—figure supplement 2C, the authors claim that no expression, or mosaic expression, of SnailA+B occurred in SnailA+B KO embryos. However, they only show the representative results of 75% of SnailA, and 49% of SnailB; how did the others look like?

The others look like the control expression. A note was added to the legend.

Also, there are faint bands on the gel running the PCR products from the KO embryos on panels A & B, suggesting that KO wasn't complete.

As we show by ISH, there is some mosaicism present in the KO embryos. Similarly, to the case for Par3 KO, we picked up a limited number of embryos out of over 2000 injected embryos, thus, the probability to have mosaicism or wild type embryos are very high. We decided then to assess the phenotypes and results by IHC and ISH.

It is not clear to us how the authors concluded that ectodermal cell fate had not changed.

We performed ISH for brachyury, chordin, and six3/6 that are well known ectodermal markers in N. vectensis. We did not observe a significant change in gene expression even though the morphology of the embryos was highly disrupted.

6) For review and visualization purposes, the figures are so small that it is challenging to see some of the most striking results at cellular resolution.

We have separated some of the main figures (now 8 figures) and their supplemental data in order to increase the size of important results. We are more than happy to make any further modification as necessary.

Please provide a representative z-stack for data from some of the experiments. For example, Figure 5D would benefit from doing this, to better see bottle cell morphology following Snail induction.

We have included Videos and ‘*.gif’ files of the z-stacks for the most representative results. We are more than happy to share more figures as necessary.

Z-stacks (*.gif’ format. It can be opened and edited using FIJI) were uploaded as supplementary files:

Figure 3E z-stack: Related to Figure 3E CRISPR phenotype (Video 1).

Figure 4 z-stack NvPar-6: Related to Figure 4 CRISPR phenotype for NvPar-6 antibody.

Figure 4 z-stack ß-catenin: Related to Figure 4 CRISPR phenotype for ß-catenin antibody (Video 2).

Figure 6D z-stack NvPar-6: Related to Figure 6D for NvPar-6 antibody (Video 5).

Figure 6D z-stack NvPar-1: Related to Figure 6D for NvPar-1 antibody (Video 6).