Bacteria Breed Multicellularity?

A single-celled relative of animals forms colonies when exposed to a bacterial product, hinting at the possible origins of multicellularity.

Aug 15, 2012
Hayley Dunning

Choanoflagellates, single-celled flagellates, have been thought of as a model for multicellularity since the 1800s because they live either as individuals or in colonies shaped like rosettes. The way colonies form is also intriguing: when daughter cells are spawned, they stick around instead of breaking up, in what looks at first glance like the dividing cells of an embryo. Now, in a paper to be published in the first issue of the upcoming open-access journal eLife, a driving factor of rosette formation has been uncovered. A sulfonolipid, produced by a bacteria that choanoflagellates eat, induces colony development, raising the possibility that bacteria were involved in the evolution of multicellular life.

"The origin of eukaryotic cells occurred in a really dense mass of bacteria," said zoologist Michael Hadfield at the University of Hawaii, who was not involved in the study. There's "just no way" that those first organisms weren't involved with bacteria in many ways beyond just feeding on them, he said. "What this new paper shows us is that there's this really fundamental connection going on here between [bacteria] and things that operate as single cells ,then start operating as colonies."

While recent genetic studies of choanoflagellates have reinforced their position as good models of multicellularity, it had proven hard to make them cooperate in the lab. In Nicole King's lab at Berkeley, however, by treating them with antibiotics—a ruthless measure to stop the choanoflagellates being overrun by bacteria—researchers found the rosettes would no longer form. Suspecting the bacteria thus had a role in rosette formation, the team tested over 60 bacterial strains that were associated with the original choanoflagellate sample, and found just one that was associated with the choanoflagellates sticking together upon dividing.

"This wasn't an autonomous decision by the choanoflagellates, but it was something prompted by their environment," said John Clardy, a molecular biologist at Harvard who helped identify the compound responsible.

The bacterium responsible was a newly discovered species (Algoriphagus machipongonensis) of the phylum Bacteroidetes, other members of which influence algal development and even the development of the immune system in animals. By feeding groups of choanoflagellates different bacterial species from the original sample, the team found that when fed solely on A. machipongonensis, the choanoflagellates produced high numbers of rosettes. This phenomenon was not observed with any of the other bacteria. The chemical mechanisms of bacterial interactions are poorly understood, however, and it took the team a while to isolate the compound responsible. It turned out to be a rare lipid, a sulpholipid they termed Rosette-Inducing Factor 1 (RIF-1). The compound appears to be remarkably potent, with just nanomoles of the stuff necessary to induce rosette formation. Intriguingly, rosettes are more efficient than individuals at capturing A. machipongonensis and consuming them, and why the bacteria would produce a compound that quickens their demise is a mystery the team are pursuing.

While choanoflagellates are not our ancestors, Hadfield is excited by the implications of the finding in a close relative to all animals. "It suggests that [to go] from single-celled animals that had to catch and eat their food to a multicellular state, bacteria were involved," he said.

Whether it is the origin of multicellular life, however, is an open question, according to Clardy. "You can talk to people who will say 'No, this is only one way,' or 'This is a red herring,’" he said. "But what I think is interesting is that this is a fabulous example of how bacteria profoundly influence the development of eukaryotes."

"It's a remarkable discovery," said Randy Schekman, a cell biologist at the University of California and the editor-in-chief of eLife. "Who would have guessed a molecule like that could be involved in colonization?" Schekman has been aware of the work in King's lab for some time, and when he heard about the results they'd achieved, he set about encouraging them to submit a paper to his new open-access venture. Convincing all of the paper's authors that they would get the appropriate exposure wasn't easy for a new journal, Schekman said, but in the end they wanted to take part in a new model of open-access publishing. (See this month's feature on the future of science publishing for more discussion of open-access publishers.) The full article will be published in eLife's first issue, to be released in a couple of months, but a pre-print is currently available on the King lab website.

R. Alegado et al., "Bacterial regulation of colony development in the closest living relatives of animals," eLife, citation to be confirmed, 2012.