Plant and animal development compared

Blogging on Peer-Reviewed Research

Since I wrote about the wacky creationist who couldn’t wrap his mind around the idea that plants and animals are related, and since I generally do a poor job of discussing that important kingdom of the plants (I admit it, I’m a metazoan bigot…but I do try to overcome my biases), I thought I’d briefly mention an older review by Elliot Meyerowitz that compares developmental processes in plants and animals. The main message is that developmental processes, the mechanisms that assemble the multicellular whole, are very different in the two groups and are non-homologous, but don’t get confused: the basic cellular processes are homologous, and there’s no doubt that we are related. The emphasis in this paper, though, is the evidence that plants and animals independently evolved multicellular developmental strategies. There is some convergence, but the tools in the toolbox are different.

First, here is a basic overview of the evolutionary history of these two groups.


Notice that we’re talking about events in deep time here: we eukaryotes (organisms with a true cell nucleus) evolved as single-celled animals 2.7 billion years ago, and we only know this because we find smears of characteristic hydrocarbons in the rocks of that age. The last common ancestor of plants and animals is estimated from molecular clock data to have lived 1.6 billion years ago — that’s a very long time ago. It’s hard to get the concept of that huge amount of time across to people, but look at the relative amounts of time: complex, multicellular animals evolved a bit more than 0.5 billion years ago, and all of the macroscopically large animal diversity we see now evolved since then; but there’s a period twice as long as that before arthropods and molluscs developed hard shells in which change was percolating in single-celled eukaryotes, laying the foundation for the current differences between orange trees and cats.

But wait…what was the plant-animal common ancestor like? Was it a single-celled organism, or did it have a more complex multicellular organization? Meyerowitz catalogs the different molecular tools that plants and animals use in development, and concludes that development evolved completely independently in the two, with a few hints of some shared elements that tell us a bit about the genetic heritage we have received from that mysterious Proterozoic cell.

Meyerowitz summarizes a few broad categories of similarities and differences in plant and animal development.

  • Pattern formation: Pattern formation is a very wide assortment of developmental processes that define regions of an embryo — the events that put a head on one end and a tail on the other, or a flower above ground and roots below ground, are examples of pattern formation. In animals, the classic example of pattern forming genes are the Hox genes, which establish regional specifications in the early embryo. Plants have similar genes, the MADS box genes, that also set out overlapping regional identities in the growing plant — but MADS boxes and Hox genes are not homologous. Animals have MADS box genes, but don’t use them in pattern formation; plants have Hox genes, but also don’t use them in pattern formation. This suggests that the last common ancestor of plants and animals was single-celled or at best colonial, and lacked any differently developed regions; when regional specialization evolved later in each lineage, different sets of transcription factors were arbitrarily used for the job.

    Another example is dorsoventral specification. We know quite a bit about how that axis is defined in animals, and it’s an elaborate process involving a TGF-α-family protein, an EGF receptor, a receptor tyrosine kinase, a Ras activated cascade, etc. Trust me, it’s detailed, and to developmental biologists with a metazoan bias, like me, the components are all familiar friends. Plants have a similar task, setting up the adaxial-abaxial axis (you’ve certainly noticed that the top and bottom sides of leaves are different), but the molecules involved are completely different, and plants lack homologs of many of them altogether. Instead, plants use REVOLUTA/PHABULOSA/PHAVOLUTA receptors and REV/PHAB/PHAV homeobox proteins, KANADI genes, and the YABBY family of transcription factors. It’s a whole new language as far as I’m concerned, but fascinating — plants use a similar logic to animals, but the pieces are alien to me.

  • Chromatin processes: Another critical step in development is the maintenance of a pattern. After pattern formation initializes the organization, cells in the embryo that descend from an early cell with a particular identity have to maintain that specification; development is a strongly hierarchical process. There are gene products that maintain the pattern of activators and repressors of gene activity and that are inherited by daughter cells in mitosis. In flies, for example, there is a gene called Enhancer of zeste, E(z), that helps lock in expression of some of the Hox genes. Plants have similar genes, and in some cases at least, they’ve been found to be homologous. Arabidopsis has a gene called CURLY LEAF (CLF) that is an E(z) homolog, and there are other chromatin-regulatory proteins that are found in both plants and animals. CLF and E(z) are clearly related genes, and they carry out similar functions in the regulatory logic, but note that while E(z) regulates a Hox gene, CLF regulates a MADS box gene.

  • Cell:cell signaling: During development, cells must communicate with each other, and they do so with molecules, ligands and receptors, that can activate signal transduction cascades inside the cell that lead to changes in metabolic activity or gene regulation. In animals, one common and important family of cell signaling receptors are the receptor tyrosine kinases (RTKs). A kinase is an enzyme that phosphorylates other proteins, attaching a phosphate group to certain amino acids. A tyrosine kinase is one that phosphorylates tyrosines. Plants don’t have any RTKs at all — they are unique to the animal (well, opisthokont) lineage. Plants certainly do have receptor kinases, and carry out cell signaling quite well, but are more likely to use a receptor serine/threonine kinase. Again, we find similar logic carried out by nonhomologous components.

    Another example are the familiar steroid hormones. Both plants and animals use steroid signaling, but animals receive that signal in each cell via members of the nuclear hormone receptor family, while plants use a receptor kinase. Steroid synthesis is common to both lineages, but each has its own way of using it as a signal.

  • Horizontal transfer: Don’t get the impression that animals have all the cool, unique, new stuff, though — that isn’t true. Plants also have their own unique flavors of proteins for which we animals have no homolog. One example are the families of ethylene receptors, a chemical plants use as a signal. These receptors resemble bacterial two-component receptors, and are thought to have aisen by horizontal transfer from cyanobacteria. This transfer event probably occurred after the divergence of plants and animals, so we missed out.

    Another example are the phytochromes, red and far-red light receptors that are similarly related to bacterial two-component receptors. We have nothing similar. What’s particularly interesting about these is that in cyanobacteria, they act as histidine kinases, but in plants, they fit into the more common plant pathways by working as serine/threonine kinases. While horizontal transfer may be a common phenomenon, retention and expansion of such genes shuffled across species also requires some modification to incorporate them into a functional pathway.

The bottom line is that plants and animals clearly arose from a common ancestor, almost certainly single-celled, and that they’ve evolved the processes that allow cells to cooperate and communicate and assemble into complex, elaborate entities with tissues and organs nearly completely independently. We don’t need to go to Mars or Betelgeuse to find aliens, they’re living side-by-side right here on planet Earth. Most importantly, if evo-devo wants to find truly general, universal principles of multicellular development, we can’t get too fixated on gene identities in metazoans (and especially not on the specifics of development in one beastie, Drosophila melanogaster) — we have to throw a wider phyletic net and distill out a bigger picture of the mechanisms involved. The important focus should be on developmental logic, rather than developmental details.

Meyerowitz EM (2002) Plants Compared to Animals: The Broadest Comparative Study of Development. Science 295(5559):1482-1485.


  1. #1 David Marjanovi?, OM
    February 17, 2008

    What happened to the 2.1 Ga old fossils of multicellular red algae? Did they get radically reinterpreted? And aren’t the 2.7 Ga old chemofossils dinosteranes, something that only dinoflagellates produce? (The dinoflagellates are nested deep inside a large tree, so most of eukaryote diversity must have been present when the first dinoflagellates appeared.)

    The horizontal transfer from cyanobacteria — is there any evidence that this didn’t happen at the same time as the origin of chloroplasts?

    Something I still don’t understand ; that single celled common ancestor must have had a very short genome.

    Not at all, why?

    but don’t these two play the same kind of role in the cell? Mitochondria uses oxygen to react with ATP to make chemical energy available to the cell, and chloroplasts use photons to do something similar?

    Yes, but chloroplasts work only when and where the sun shines. The stem and roots (and the epidermis) have no choice but to breathe, and at night the whole plant does so.

    This is the whole point of making sugar in the chloroplasts.

    Where do fungi branch off on that diagram, before or after the plant/animal split?

    Long after. Fungi and animals are very close relatives. (Together they are called Opisthokonta because they — originally — have a single cilium that inserts at the back end of the cell.)

  2. #2 David Marjanovi?, OM
    February 17, 2008

    The thylakoids are simply budded-off infoldings of the inner membrane. The inner membrane, like that of mitochondria (which also has lots of infoldings), is the chloroplast’s own membrane, and the outer one ultimately results from the endocytosis event that led to the origin of chloroplasts.

    The chloroplasts of the glaucophytes, called cyanelles, have a bacterial cell wall in the expected place: between the two membranes.