Tiny sea creatures upend notion of how animals' nervous systems evolved

Sweeping study of sea creatures suggests wild deviations over evolutionary time.

Search for this author in:

Xenoturbella bocki

Shown here is beautiful image of Xenoturbella bocki, a firm favourite in the scientific community.Credit: Ulf Jondelius

A study of some of the world’s most obscure marine life suggests that the central nervous system evolved independently several times — not just once, as previously thought1.

The invertebrates in question belong to families scattered throughout the animal evolutionary tree, and they display a diversity of central nerve cord architectures. The creatures also activate genes involved with nervous system development in other, well-studied animals — but they often do it in non-neural ways, report the authors of the paper, published on 13 December in Nature.

“This puts a stake in the heart of the idea of an ancestor with a central nerve cord,” says Greg Wray, an evolutionary developmental biologist at Duke University in Durham, North Carolina. “That opens up a lot of questions we don’t have answers to — like, if central nerve cords evolved independently in different lineages, why do they have so many similarities?”

In 1875, German zoologist Anton Dohrn noted anatomical similarities between the central nerve cord that runs length-wise through the bodies of annelids — a group of invertebrates that includes earthworms — and the nerve cord in the spine of vertebrates. He proposed that the groups’ ancient common ancestor had a nerve cord that ran along its belly-side, as seen in annelids. He also suggested that this cord flipped to the back of the body in a more recent animal that gave rise to all vertebrates.

More than a century later, evolutionary developmental biologists revisited Dohrn’s theory when they discovered that the same genes involved in the development of vertebrates' central nerve cords are also activated in the nerve cord of the fly Drosophila melanogaster and of the marine annelid Platynereis dumerilii2. Similar gene expression underscored the concept that the cords could be traced back to a common ancestor.

Brachiopod larva, SEM.

SEM of an early brachiopod lava.Credit: Jürgen Berger and Andreas Hejnol

Developmental debate

But researchers questioned the theory in 2006, after looking at the expression of a suite of those genes — including one named bmp — in acorn worms3. They found that bmp is activated in these animals early in their development — well before they form two nerve cords that run along the sides of their bodies. The scientists suggested that bmp helps to provide coordinates for cells in the developing embryo. But rather than do away with the idea that bmp unites the nerve cords of disparate species, many biologists suggested that acorn worms might be an exception in using the gene in a different way. After all, they had unusual, dual nerve cords.

Andreas Hejnol, an evolutionary developmental biologist at the Sars International Centre for Marine Molecular Biology in Bergen, Norway, and senior author of the Nature paper, was fascinated. “I thought, you should not call an animal weird,” he says. “Let the animals tell you who is weird.”

In search of creatures with diverse nervous systems, Hejnol’s team explored fjords in Sweden and Norway by boat. They sifted through sludge drudged up from the sea floor, and probed the guts of sea cucumbers to find parasites buried within. The scientists also scoured the rocky shores of islands off Washington state.

Some of the tiniest worms the team collected belong to an ancient lineage in the animal evolutionary tree called Xenacoelomorpha, and they possessed a plethora of nervous systems. For instance, Xenoturbella bocki has no central nerve cord, but rather a net of nerves similar to those in jellyfish; Isodiametra pulchra has eight nerve cords; and Meara stichopi has a nerve cord running along its back, as vertebrates do.

Researchers collect specimens in Norway.

Researchers collecting nemerteans (ribbon worms) in Bergen, Norway.Credit: Eivind Senneset.

Cords of all forms

As in the acorn worms, these itty-bitty worms activated bmp before nerve cords formed, early in embryonic development. Moreover, Hejnol blocked the protein pathway and found that the animals’ nerve cords developed regardless. The result suggests that they are constructing their nerve cords differently than mice, flies and other well-studied animals.

In lampshells — sea creatures named after the shape of their protective coverings — Hejnol’s group found that other genes previously associated with the central nerve cord switched on as the lampshells developed, even though the creatures have no central nerve cord. The discordance continued in the wheel-bearers, the nemerteans and other odd animals.

Hejnol concludes that genes found to underlie the central nerve cord in vertebrates, flies and some annelids functioned differently in an early ancestor, and were instead integrated into the nervous system at different points in time as disparate animal lineages evolved central nerve cords of their own.

However, Linda Holland, an evolutionary biologist at the University of California, San Diego, is not ready to make such a pronouncement. She says that the creatures Hejnol studied might instead deviate from the ancestral condition. It may be impossible to know what that ancestor looked like, she adds, because more than 500 million years of evolution has passed since these lineages split from one another. “It can be a little uncomfy,” she says, “but I’m sitting on the fence.”

Wray is ready to dump Dohrn’s single origin scenario, but he suspects the truth is more nuanced. “There are probably some components [of the central nerve cord] that are maintained from ancient evolutionary times," he says, "and some that are layered on top of that.”

Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up


  1. 1.

    Martín-Durán, J. M. et al. Nature (2017).

  2. 2.

    Holland, L. Z. et al. Evo. Devo. 4, 27 (2013).

  3. 3.

    Lowe, Christopher J. et al. PLoS Biol. 4, e291 (2006).

Download references