Life-stage-specific specialities in the cell atlases of the Clytia hemisphaerica planula and medusa
Curation statements for this article:-
Curated by eLife
eLife Assessment
This study provides a useful single-cell atlas of the Clytia hemisphaerica planula, complemented by an updated medusa dataset, ultrastructural analyses and in situ validation of expression patterns. The cross-stage comparison and cluster-similarity framework offer a promising basis for investigating cellular diversification across the life cycle. The evidence has the potential to be convincing, but is currently incomplete due to documentation deficits and insufficient cross-referencing with prior work. It should be relatively easy to address these issues, which should make this a valued resource for cnidarian researchers and for colleagues studying cell-type evolution.
This article has been Reviewed by the following groups
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
- Evaluated articles (eLife)
Abstract
Jellyfish have complex life-cycles, but there has been limited exploration of how this is achieved at the cellular level. We used single-cell transcriptomics to assemble a cell atlas for the planula larva of Clytia hemisphaerica , and compared it to an updated cell atlas for the medusa (jellyfish) stage. The cells of the planula fell into the same broad categories as for the medusa: ectoderm, gastroderm, interstitial cells (i-cells), nematocytes (stinging cells), neurons and secretory cells. Although the planula cells generally showed less diversity than medusae within each category, cells with specialized features unique to their stage could be distinguished by their transcriptional profiles as well as by ultrastructure. Some planula-specific types were identified: aboral secretory cells involved in settlement, and a cell type attributed a role in immunity or post-metamorphic theca production. Distinct transcriptome profiles within different regions of the ciliated planula ectoderm reflected different post-metamorphosis fates of domains along the oral-aboral axis. Inspection of the cell clusters showing significant similarity of marker genes between planula and medusa, and inference of similarity using a statistical model of marker gene presence/absence, revealed correspondences between families of cells from planula and medusa rather than precise cell identities.
Article activity feed
-
eLife Assessment
This study provides a useful single-cell atlas of the Clytia hemisphaerica planula, complemented by an updated medusa dataset, ultrastructural analyses and in situ validation of expression patterns. The cross-stage comparison and cluster-similarity framework offer a promising basis for investigating cellular diversification across the life cycle. The evidence has the potential to be convincing, but is currently incomplete due to documentation deficits and insufficient cross-referencing with prior work. It should be relatively easy to address these issues, which should make this a valued resource for cnidarian researchers and for colleagues studying cell-type evolution.
-
Reviewer #1 (Public review):
Summary:
The manuscript further explores the single-cell atlas of Clytia hemisphaerica by incorporating the planula larva. It compares the cell clusters with the previously established atlas of the medusa. It identifies similarities and differences between the two life stages.
Strengths:
The manuscript provides an important set of single-cell data that have not been assessed previously: the Clytia planula. The data is further supplemented with high-quality EM-based histology and an extensive in situ hybridisation of selected genes.
Weaknesses:
The detailed analysis does not go deep into the comparison between stages, nor does it provide an analysis of genes within the clusters; it could be described as remaining overall rather superficial.
-
Reviewer #2 (Public review):
Summary:
The generation of alternate stages in the life cycle of a single species requires vast remodeling of the cellular complement of the individual during metamorphosis from one stage to another. In this paper, the authors provide a detailed description of single-cell RNA-seq data derived from the planula stage of the hydrozoan model Clytia hemispherica and compare this to an expanded dataset from the medusa stage to assess changes in transcriptomic identity of cell types between these two phases of the life cycle. The paper further includes valuable TEM data illustrating fine anatomy of cell types present at both stages investigated, and documentation of the retention of epithelial polarity from the planula through to the polyp stage, using a reporter line.
Strengths:
The study provides a solid and …
Reviewer #2 (Public review):
Summary:
The generation of alternate stages in the life cycle of a single species requires vast remodeling of the cellular complement of the individual during metamorphosis from one stage to another. In this paper, the authors provide a detailed description of single-cell RNA-seq data derived from the planula stage of the hydrozoan model Clytia hemispherica and compare this to an expanded dataset from the medusa stage to assess changes in transcriptomic identity of cell types between these two phases of the life cycle. The paper further includes valuable TEM data illustrating fine anatomy of cell types present at both stages investigated, and documentation of the retention of epithelial polarity from the planula through to the polyp stage, using a reporter line.
Strengths:
The study provides a solid and convincing transcriptomic characterization of planula cell types (including in situ validations and a planula-to-polyp mapping of epithelial polarity), and introduces a potentially valuable method for evaluating cluster similarity.
Weaknesses:
The work suffers from insufficient documentation of methodological approaches and missing code, lack of clarity regarding clustering resolution and nomenclature (thereby hindering cross-referencing with prior papers), and unclear plans for public, fully annotated data release.
Full Review:
The single-cell transcriptomic data analyzed include both previously published and newly generated data: two additional medusa libraries and two additional planula libraries were generated and integrated with the data from https://doi.org/10.1126/sciadv.abh1683, and https://doi.org/10.1126/sciadv.adv1159. The original release of the planula dataset in their 2025 Science Advances paper did not include analyses of all cell types. Here the authors provide this analysis for the planula stage. However, as both the number of clusters and the nomenclature of the clusters changed, this leads to some confusion and inability to cross-reference the two papers. There is no explanation given for the re-processing of the planula dataset in the current paper, and the fact that only some of the data is new is buried in the supplement, which is not referenced in the main document, while the text within the main article suggests that the entire dataset is new. The fact that the dataset in the current analyses contains fewer cells than presented in their Science Advances paper further adds to this confusion. The current paper would benefit from greater transparency in the origin of the data analyzed.
The authors do try to apply the same nomenclature for the updated medusa dataset that is present in their 2021 Science paper. For example, the previously identified 'bioluminescent cells' are identified as 'gas-m8'. A look-up table that has all of the cluster id's cross-referenced would be useful (i.e. new: 8 = gas-m8 = previous: 28 = BC = "Tentacle GFP cells"). The inability to easily cross-compare with the published data is a major weakness of the current work and would benefit greatly from consistency between the three papers. Indeed, the clustering resolution is quite different across all three papers, and the current work does not adequately address how the clustering resolution was selected here. As an updated atlas, one would expect the entire transcriptomic diversity to be included here, so that the previous work can be transferred to the updated genomic mapping resource used in the current work. Nonetheless, presenting a unified nomenclature for moving forward would benefit the community as a whole and would increase the impact of the current work substantially.
The paper also includes new TEM data of the planula cell types. The authors attempt to correlate transcriptomic profiles with these anatomical data through in situ hybridizations that provide spatial distribution of the profiles. While the TEM data are valuable to catalog the presence of cells with different morphologies within the planula, the association with the transcriptomic profiles is somewhat speculative. These valuable anatomical data should be provided at a high enough resolution to zoom in and see the details, and further description could be provided. For example, the paper states that vacuolated cells are characteristic of the basal gastrodermal cells adjacent to the mesoglea; please identify the vacuoles in Figure 4f/h for the reader.
A novel method for reconstructing cluster similarity relationships is applied to grouping clusters into cell categories within the same life cycle stage, and also for matching cell types between stages. This is a valuable contribution to the field that is worthy of further evaluation. This is, however, difficult, as the methods for which DESeq2 was applied ("see code for details") are not present in the provided code, nor is it adequately described how the "binary matrix of marker gene presence/absence" was constructed. Similarly, there are additional details of other parts of the data analysis that are missing from the provided code, and the provided supplementary material is not referenced in the main document. More rigorous documentation of the methods is warranted.
The description of the transcriptomic profiles present in the planula is solid, and the attempt to associate these profiles with anatomic locations and putative morphology provides a foundation onto which further studies can be developed. Mapping of the planula ectoderm through to the polyp stage is also an important step forward in characterizing the life cycle, and the evidence for the retention of the oral/aboral ectodermal axis is convincing. The paper falls short in describing the updated medusa dataset and could benefit from a minor restructuring of the paper. Introducing the new medusa data only after the planula dataset is fully described would mediate the shallower treatment of the updated medusa dataset, where only 22 of the original 36 transcriptomic states are recovered. In this way, the focus will shift onto the cross-life cycle stage comparisons, and it could be argued that the lower resolution of the medusa dataset is justified in order to simplify the comparisons.
It will be essential that the datasets that are presented in this work be made available for public exploration in a fully annotated format. It is currently unclear how the authors intend to do this; however, there are many repositories available for this. The UCSC Cell Browser hosted at cells.ucsc.edu is one very good option if the authors do not wish to develop an interactive tool themselves. It is imperative that the gene annotations which correspond to the dataset, and the cluster annotations that are presented in this paper, are available and easily connected to the released dataset.
-
