Evolution: Tracing the history of cell types

A study of sea urchin and sea star larvae paves the way for understanding how cell types evolve and give rise to novel morphologies.
  1. Antonia Grausgruber
  2. Roger Revilla-i-Domingo  Is a corresponding author
  1. Department of Neuroscience and Developmental Biology, University of Vienna, Austria
  2. Research Platform Single Cell Regulation of Stem Cells, University of Vienna, Austria
  3. Max Perutz Labs, University of Vienna, Austria

For decades scientists have been fascinated with how evolution gave rise to the wide range of shapes and sizes that exist in the animal kingdom today. At the heart of this morphological variation are the different cell types that make up an organism, some of which will be unique to certain species (Arendt et al., 2016). However, how new cell types emerged over the course of evolution is still hotly debated.

A group of animals known as echinoderms – which includes sea urchins, sea stars and brittle stars among others – offer a fascinating example of morphological novelty. Although their larvae share many morphological features, there are also some clear distinctions, making them a useful system for studying how specific cell types evolved. For instance, the two most well-studied echinoderms, sea stars and sea urchins, develop in a remarkably similar way. However, despite this, sea urchins have two cell types that are not found in sea stars (Figure 1A): skeletogenic mesenchyme cells, which make the mineral skeleton that gives sea urchin larvae their distinct shape, and pigmented cells that reside in the outer epithelium and are recruited to the immune system during an infection (Decker and Lennarz, 1988; Gustafson and Wolpert, 1967; Buckley and Rast, 2017). Now, in eLife, Veronica Hinman and colleagues from Carnegie Mellon University – including Anne Meyer as first author – report new insights in to how these two cell types emerged over the course of evolution (Meyer et al., 2023).

Morphological novelties in the sea urchin larva.

(A) Schematic representation of the adult (top) and larval (bottom) forms of sea stars and sea urchins. Two larval cell types had been thought to be unique to the sea urchin: the skeletogenic mesenchyme cells that make the skeleton (red), and the pigment cells embedded in the larval epithelium (purple). Meyer et al. found that the skeletogenic mesenchyme cells express similar genes to the right coelomic pouch of the sea star larva (blue), and the pigment cells are similar to immune and neural cells of the sea star larva (not shown). (B) These cells were found to share regulatory mechanisms (represented by different coloured clockwork wheels). Pigment cells activate similar regulatory components as immune (orange wheel) and neural (yellow wheel) cells in sea stars. Unknown components (grey wheels) may control the interplay between the immune and neural functions of these cells. The right coelomic pouch of the sea star turns on regulatory components required to produce mesenchymal cells (green wheels) – epithelial to mesenchymal transition (EMT), cell migration, and tubulogenesis – which are also activated in skeletogenic mesenchyme cells in sea urchins. This suggests that skeletogenic mesenchyme cells and cells of the right coelomic pouch evolved from the same ancestor. Later in evolution, the skeletogenic mesenchyme cells acquired a skeletogenic program (red wheel), while the right coelomic pouch cells might have acquired other currently unknown programs (blue wheel).

Image credit: Antonia Grausgruber.

Cell types are typically defined based on their morphology, function, and where and when in the embryo they originated. But Meyer et al. propose that analysing these attributes alone is not sufficient for tracing how a cell type evolved. Instead, the team focused on the inner machinery that regulates all attributes of every cell. They did this by employing a technique called single-nuclei RNA sequencing, which reveals the genes expressed in every cell of an organism. This information can then be used to determine which cells activate similar regulatory mechanisms and therefore might be evolutionary related (Marioni and Arendt, 2017; Tanay and Sebé-Pedrós, 2021).

Meyer et al. used this powerful method to see how cells in the larva of a sea star compare to previously published sequencing data from the larva of a sea urchin (Foster et al., 2020). This revealed a very different picture than expected: both cell types that were thought to be unique to sea urchins appeared to have counterparts in the sea star larva (Figure 1A and B).

The skeletogenic mesenchyme cells of the developing sea urchin were similar to cells of the right coelomic pouch in the sea star which buds off the top of the embryonic structure that gives rise to the gut (Child, 1941). Many of the genes they shared are known to regulate the developmental processes required to generate mesenchymal cells, including epithelial to mesenchymal transition (EMT), cell migration and tubulogenesis. This suggests that skeletogenic mesenchyme cells and cells in the right coelomic pouch evolved from a common ancestor, which also possessed these regulatory mechanisms (Figure 1B). It is likely that these cells then evolutionarily diverged to activate the sets of genes responsible for building the skeleton in sea urchins (Gao and Davidson, 2008).

The pigment cells of the sea urchin were also found to be similar to immune and neural cell types in the sea star. Based on the genes shared between them, Meyer et al. propose that pigment cells have a dual role (immune and neural) in sea urchin larvae, while these functions are exerted by two separate cell types in sea stars (Figure 1B). Although a specific evolutionary history for these cell types is not provided in this study, these results support previous findings suggesting that immune and neural cells are evolutionarily related (Klimovich and Bosch, 2018).

A number of questions still remain about the evolutionary history of skeletogenic mesenchyme and pigment cells. For example, while the pigment cells seem to be unique to sea urchins, a larval skeleton is also present in other echinoderms, such as brittle stars. Extending the approach used by Meyer et al. to more echinoderms will provide a deeper understanding of how the cell types that give rise to the larval skeleton emerged. In addition, unravelling the networks of genes that control the development of skeletogenic mesenchyme and pigment cells in various echinoderms will help reveal the precise events that led to these evolutionary novelties. The findings of Meyer et al. provide a very fertile ground from which to address these questions, and more broadly how the enormous diversity of animal forms arose during evolution.

References

Article and author information

Author details

  1. Antonia Grausgruber

    Antonia Grausgruber is in the Department of Neuroscience and Developmental Biology, and the Research Platform Single Cell Regulation of Stem Cells, University of Vienna, Vienna, Austria

    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0009-0005-9874-2882
  2. Roger Revilla-i-Domingo

    Roger Revilla-i-Domingo is in the Department of Neuroscience and Developmental Biology, and the Max Perutz Labs, University of Vienna, Vienna, Austria

    For correspondence
    roger.revilla-i-domingo@univie.ac.at
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-7943-5776

Publication history

  1. Version of Record published:

Copyright

© 2023, Grausgruber and Revilla-i-Domingo

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Antonia Grausgruber
  2. Roger Revilla-i-Domingo
(2023)
Evolution: Tracing the history of cell types
eLife 12:e90447.
https://doi.org/10.7554/eLife.90447

Further reading

    1. Developmental Biology
    2. Stem Cells and Regenerative Medicine
    Olivia B Taylor, Nicholas DeGroff ... Andy J Fischer
    Research Article

    The purpose of these studies is to investigate how Sphingosine-1-phosphate (S1P) signaling regulates glial phenotype, dedifferentiation of Müller glia (MG), reprogramming into proliferating MG-derived progenitor cells (MGPCs), and neuronal differentiation of the progeny of MGPCs in the chick retina. We found that S1P-related genes are highly expressed by retinal neurons and glia, and levels of expression were dynamically regulated following retinal damage. Drug treatments that activate S1P receptor 1 (S1PR1) or increase levels of S1P suppressed the formation of MGPCs. Conversely, treatments that inhibit S1PR1 or decrease levels of S1P stimulated the formation of MGPCs. Inhibition of S1P receptors or S1P synthesis significantly enhanced the neuronal differentiation of the progeny of MGPCs. We report that S1P-related gene expression in MG is modulated by microglia and inhibition of S1P receptors or S1P synthesis partially rescues the loss of MGPC formation in damaged retinas missing microglia. Finally, we show that TGFβ/Smad3 signaling in the resting retina maintains S1PR1 expression in MG. We conclude that the S1P signaling is dynamically regulated in MG and MGPCs in the chick retina, and activation of S1P signaling depends, in part, on signals produced by reactive microglia.

    1. Developmental Biology
    Kayleigh Bozon, Hartmut Cuny ... Sally L Dunwoodie
    Research Article

    Congenital malformations can originate from numerous genetic or non-genetic factors but in most cases the causes are unknown. Genetic disruption of nicotinamide adenine dinucleotide (NAD) de novo synthesis causes multiple malformations, collectively termed Congenital NAD Deficiency Disorder (CNDD), highlighting the necessity of this pathway during embryogenesis. Previous work in mice shows that NAD deficiency perturbs embryonic development specifically when organs are forming. While the pathway is predominantly active in the liver postnatally, the site of activity prior to and during organogenesis is unknown. Here, we used a mouse model of human CNDD and assessed pathway functionality in embryonic livers and extraembryonic tissues via gene expression, enzyme activity and metabolic analyses. We found that the extra-embryonic visceral yolk sac endoderm exclusively synthesises NAD de novo during early organogenesis before the embryonic liver takes over this function. Under CNDD-inducing conditions, visceral yolk sacs had reduced NAD levels and altered NAD-related metabolic profiles, affecting embryo metabolism. Expression of requisite pathway genes is conserved in the equivalent yolk sac cell type in humans. Our findings show that visceral yolk sac-mediated NAD de novo synthesis activity is essential for mouse embryogenesis and its perturbation causes CNDD. As mouse and human yolk sacs are functionally homologous, our data improve the understanding of human congenital malformation causation.