Menu

Close

A unicellular relative of animals generates an epithelium-like cell layer by actomyosin-dependent cellularization

Omaya Dudin, Andrej Ondracka, Xavier Grau-Bové, Arthur A.B. Haraldsen, Atsushi Toyoda, Hiroshi Suga, Jon Bråte, Iñaki Ruiz-Trillo

Preprint posted on February 28, 2019 https://www.biorxiv.org/content/10.1101/563726v1

(Transiently) Comfortable in its own “skin”: formation of epithelium-like multicellular structures in a unicellular organism through conserved actomyosin-dependent mechanisms.

Selected by Paul Gerald L. Sanchez and Stefano Vianello

Background

In multicellular organisms, cells arrange over time and space to give rise to tissues with distinct architectures and functional properties. Cells will arrange themselves as ordered arrays in epithelial tissues, they will be loose and sparse in connective tissues, they will give rise to even more distinctive architectures in muscle and nervous tissues. Of all these forms of organisations however, epithelia seem to hold a special place in biology. During early embryonic development of mammalian species, it is indeed as an epithelium that cells first arrange themselves when forming the blastocyst. It is still an epithelium that acts as the substrate of gastrulation in chicken, mouse, and humans. It is again to an epithelium that cells apparently default to when even just seeded within a matrix. Epithelia can fold, form sheets and tubes, and they can support morphogenesis. Epithelia have a barrier function and a polarity: they compartmentalise space and they can distinguish and generate differences between these compartments.   

Motivating the work highlighted by this preLight, epithelia have been suggested to be the first form of multicellular organisation to arise during evolution. While their ubiquitous presence across all animals clearly makes them a characteristic feature of this group, the discovery of polarized animal-like epithelia in social amoebae raised the possibility that these structures might even pre-date animals, and as such reflect an even more ancestral cellular programme. Yet, the evolutionary branches separating animals from slime moulds are also populated by unicellular life forms: epithelial organisation might thus just be the result of convergent independent evolution of epithelial-like organisation from a non epithelial ancestor.

The authors here investigate an intermediate branch of the evolutionary space separating animals from slime-moulds (i.e. the ichthyosporeans branch), and beautifully describe here too an epithelial-like life stage. Even more interestingly, the cellularization process giving rise to such epithelium occurs via animal-like mechanisms, strengthening evolutionary models that see epithelial structures as a basal characteristic of not only animals but of all unikonts.

 

Life cycle of S. arctica, and epithelial structure
FIGURE A: FIGURE 1B AND 2B (PANEL IV, MIDDLE SECTION) OF THE PREPRINT
Life cycle of Sphaeroforma arctica. Inset: labeling for nucleus (DAPI, cyan) and actin (phalloidin, green), for a sample fixed just before the flip event, showing polarized nuclear localization and cellularization.

 

Key findings

The authors previously found that the single-cell ichthyosporean Sphaeroforma arctica transits through a multicellular stage [Ondracka et al, 2018]. They find that this stage displays features typical of an epithelium (Figure A). Such a transient epithelial-like structure originates by cellularization of a coenocyte, a shared multinucleated cytoplasm generated by rounds of cell division without cytokinesis.

By using a combination of membrane and cytoskeleton marking, live imaging, and RNA sequencing of cultured organisms, the authors find that this cellularization process occurs through animal-like mechanisms. Specifically, they find that it occurs through the stepwise assembly and deployment of a cortical network of actin and myosin (Figure B):

  1. nucleation of actin at the cortex,
  2. formation of filaments,
  3. myosin-dependent crosslinking,
  4. membrane invagination to compartmentalise the underlying cytoplasm

Concomitantly, the organism undergoes a drastic transcriptional shift to genes involved in cell-cell and cell-matrix interactions, as well as in cytoskeletal components. Morphologically, and mechanistically, the process is conserved with what is seen in animal models of cellularization (e.g. in Drosophila); yet also reveals ichthyosporean-specific features such as the role of transcriptional changes.

 

Proposed model of cellularization
FIGURE B: FIGURE 6D OF THE PREPRINT
Summary of actomyosin-dependent cellularization in Sphaeroforma arctica.

 

Significance

This study identifies animal-like mechanisms of epithelial formation in yet another branch of the evolutionary tree. That is, in addition to that of the slime moulds, where the identification of catenin-based epithelial structures prompted hypothesis of a pre-animal emergence of epithelial programmes [Dickinson et al, 2012]. The S. arctica studied here (ichthyosporean) belongs to a lineage even closer to the animal branchpoint. While previous descriptions in slime moulds could have just been the result of independent evolution in slime moulds and animals, finding conserved molecular mechanisms in ichthyosporeans strengthens the hypothesis of common descent from an ancestor that did already have epithelial-like life stages. Such epithelial programmes would have been maintained in the animal, ichthyosporean, and slime-mould lineages, and lost in the many exclusively unicellular life-forms observed in between these branches.

We particularly appreciated how elegant the experimental design is, considering the limited tools available when studying an emerging model system. The authors maximized the amount of biological insight obtained from membrane/cytoskeleton labelling and targeted small molecule inhibition of different steps of cytoskeletal organisation. Of notice is also the RNA sequencing data collection, aligned to a genome re-assembled by the authors themselves!

 

Open questions

  1. You seem always very careful about talking about “epithelium-like” tissue, and not “epithelium”. Could you elaborate on this distinction? What features do you see as still missing to define this as an epithelium?
  2. Could you elaborate on the differences you see when comparing to Drosophila? What do you think these reveal about the requirements and evolutionary strategies of cellularization?
  3. What tools would you like to have available to prove the actual involvement of catenins? And to test the function of the epithelium?
  4. Animal epithelia rarely form through cellularization of a coenocyte. Could you elaborate on that? How widespread is epithelialisation (rather than cellularization) outside of the animal group?
  5. Kinesin 2 is one of the few microtubule motors that also undergoes upregulation upon cellularization. Do you have any ideas on whether it contributes to membrane folding too?
  6. Coenocyte sizes seem to vary during growth. Do bigger coenocytes take shorter time to burst after the flip event, explaining the observed variability? Do they release more/bigger cells?

 

Further reading

A comprehensive review on epithelia and their function:

  • Rodriguez-Boulan, Enrique, and Ian G. Macara. “Organization and execution of the epithelial polarity programme.” Nature Reviews Molecular Cell Biology 15.4 (2014): 225.

Overview of the current evolutionary considerations on the origin of epithelia:

  • Dickinson, Daniel J., W. James Nelson, and William I. Weis. “An epithelial tissue in Dictyostelium challenges the traditional origin of metazoan multicellularity.” Bioessays 34.10 (2012): 833-840.

A previous paper by the same lab as this preprint:

  • Ondracka, Andrej, Omaya Dudin, and Iñaki Ruiz-Trillo. “Decoupling of Nuclear Division Cycles and Cell Size during the Coenocytic Growth of the Ichthyosporean Sphaeroforma arctica.” Current Biology 28.12 (2018): 1964-1969.

An article on the root as a polarized epithelium in plants:

  • Maizel, Alexis. “Plant Biology: The Making of an Epithelium.” Current Biology 28.17 (2018): R931-951.

Tags: actomyosin, cellularization, coenocyte, epithelia, ichthyosporean, multicellularity

Posted on: 5th April 2019

Read preprint (4 votes)




  • Author's response

    Andrej Ondracka, Omaya Dudin, and Iñaki Ruiz-Trillo shared

    1. You seem always very careful about talking about “epithelium-like” tissue, and not “epithelium”. Could you elaborate on this distinction? What features do you see as still missing to define this as an epithelium?

    A key feature of epithelia is the capacity to serve as a protective, polarized barrier. Here, despite demonstrating the morphological resemblance between the epithelium-like structure in arctica and polarized epithelia in animals, we still do not know if it serves the same function of protective layer. For this reason, we prefer to call it epithelium-like rather than an epithelium.

    2. Could you elaborate on the differences you see when comparing to Drosophila? What do you think these reveal about the requirements and evolutionary strategies of cellularization?

    The cellularization process in arctica and Drospohila embryo are morphologically remarkably similar. Quantitative differences can be observed – the Drosophila blastoderm is much bigger than the Sphaeroforma coenocyte, and cellularization occurs much more rapidly in Drosophila. One important difference worth mentioning is that in S. arctica, cellularization is concomitant with transcriptional activation of thousands of genes, whereas in Drosophila, cellularization depends mostly on maternally deposited mRNA, as not much transcription is observed during that stage of embryonic development. Another key difference is that some of the typical markers that characterize epithelial tissue – in particular, cadherins – that also play a role in the cellular blastoderm upon cellularization in Drosophila, are absent in S. arctica.

    Finally, a fundamental difference between both species is what cellularization leads to. In S. arctica, an organism that lives a unicellular life style, it is the means to proliferate and produce new cells that then restart the new cycle, whereas in Drosophila it is a part of the embryonic development. At the moment we can only speculate why such a proliferative life cycle, as opposed to binary cell division, evolved in ichthyosporeans, as very little is known about the ichthyosporean life style in its natural environment.

    3. What tools would you like to have available to prove the actual involvement of catenins? And to test the function of the epithelium?

    At the moment, in arctica, we lack any kind of genetic tools. The most direct experiment would be to delete the catenins (and other polarity proteins) in S. arctica and study the phenotype, although at the moment we are far from being able to do this experiment.

    4. Animal epithelia rarely form through cellularization of a coenocyte. Could you elaborate on that? How widespread is epithelialisation (rather than cellularization) outside of the animal group?

    Actually, the cellular blastoderm, the structure that results from cellularization of the embryo in Drosophila, is seen as an epithelium. However, as you mention, other animal epithelia can arise through very different processes.

    Up until now, to our knowledge, epithelial structures outside animals have only been observed in Dictyostelium, and now in Sphaeroforma arctica. Importantly, we believe that there are critical differences between the epithelial structures observed in Dictyostelium and S. arctica. In particular, in Dictyostelium, the epithelium arises through cell aggregation, whereas in S. arctica, this process is clonal.

    5. Kinesin 2 is one of the few microtubule motors that also undergoes upregulation upon cellularization. Do you have any ideas on whether it contributes to membrane folding too?

    This is a very interesting idea. Our results suggest that microtubules play a role in spatially organizing the nuclei but are not strictly essential for membrane invaginations. Based on its expression pattern, we hypothesize that Kinesin 2 might play a role during cellularization. However, without genetic tools, at the moment we cannot tell what its role might be.

    6. Coenocyte sizes seem to vary during growth. Do bigger coenocytes take shorter time to burst after the flip event, explaining the observed variability? Do they release more/bigger cells?

    This, interestingly, does not seem to be the case. We have shown in our previous work that nuclear division cycles within the coenocyte seem to have very regular timing and independent of size. Due to this, despite differences in size, coenocytes contain roughly the same number of nuclei at cellularization. It is true that there is some variability in the duration of cellularization itself, especially of the post-flip stage, but we do not think that the coenocyte size would be likely to explain the duration of cellularization.

    Have your say

    Your email address will not be published. Required fields are marked *

    This site uses Akismet to reduce spam. Learn how your comment data is processed.

    Sign up to customise the site to your preferences and to receive alerts

    Register here
    Close