Seen through a microscope, the hairy, slipper-shaped aquatic microbe Paramecium bursaria often looks as if it is bursting at the seams with tiny green marbles. Yet the verdant spheres are a different organism altogether: Chlorella, an alga that occasionally takes refuge within the confines of the paramecium’s cushy cell membrane. Each species can survive on its own, but the two frequently and repeatedly engage in an endosymbiotic partnership. In exchange for providing defense and nutrients, the paramecium demands that its algal tenants share the food they photosynthesize.

The conventional wisdom is that a paramecium refrains from digesting the algae only because it would lose out on the sugars that they create. But new work posted to the biorxiv.org preprint server suggests that the algae may also be protected by a stabilizing fail-safe system: If a ravenous paramecium makes a meal of its resident, digested bits of the algae’s RNA may interfere with the host’s ability to grow and reproduce. Understanding the dynamics between paramecia and Chlorella could yield insights into innumerable other endosymbioses found in nature and may even offer hints about what helped sustain the original endosymbiosis that is thought to have produced eukaryotes (complex cells with organelles).

“We want to understand what are the universal principles that govern interactions between eukaryotic hosts and eukaryotic endosymbionts,” said John Archibald, a biochemist at Dalhousie University who was not involved in the study.

Even though mutualistic relationships benefit both parties, it’s wrong to think of them as inherently stable, said Thomas Richards, a professor of evolutionary genomics at the University of Oxford and a senior author of the study. “They’re always trying to get one over each other,” he said. “Because of that, it’s hard to imagine you can get long-term stability.”

For example, research has shown that paramecia digest their endosymbionts in darkness, where the algae are unable to photosynthesize and appease their hosts with sugar. “The symbiont must also have something in its arsenal that is maintaining the stability as conditions fluctuate — as the context changes,” said Ben Jenkins, a postdoc at Oxford and the first author of the study. “There has to be this homeostasis that keeps it on an even keel.”

Jenkins suspected that this stabilizing mechanism might involve the organisms’ RNA because his colleagues had noticed many regions of strong similarity between host and symbiont transcripts floating around in the cytoplasm. He wondered if there might be some consequence of the confusion between host and symbiont RNA. “If we can’t tell those transcripts apart, there’s a good chance the host also wouldn’t be able to tell the difference between the transcripts,” he said.

To find out if that was the case, Jenkins and his colleagues first administered an antibiotic that disabled the algae and led the paramecia to digest their now useless residents. When the algae are digested, their cellular guts — proteins, genetic material, organelles and all — spill out into their host’s cytoplasm, where they are slowly broken apart by enzymes.

Without their algal symbionts, the number of paramecia in the samples dropped dramatically, as expected. But Jenkins and his colleagues noticed that something else was afoot when they turned off the paramecia’s system for RNA interference (RNAi) — a process that silences gene regulation and is thought to have arisen in part to combat foreign genetic material from viruses and other invaders. The modified paramecia declined in population far less than those with intact RNAi systems.

Follow-up tests showed that the free-floating algal RNA most similar to bits of its host’s genome likely set off a signaling cascade within the paramecia. The cascade communicated that the cell was becoming overrun with that genetic information, possibly indicating the presence of a virus or some other invader. This caused the cell to slow down the expression of the homologous genes and make fewer of the proteins they encode.

“[The model system] is the equivalent of a time machine — you can use it to go back in time and look at a nascent symbiotic interaction,” said Archibald. “They uncovered the language of symbiosis at the finest scale.”

Hailing Jin, a microbiologist and geneticist at the University of California, Riverside who was not involved in the study, described the study as exciting and said that it fit in with a larger body of research in the past decade stressing the importance of cells’ RNAi machinery. Only in the last five or so years, Jin said, have scientists turned their attention to investigating RNAi interactions between different organisms, species and even kingdoms.

Jenkins noted that what makes this RNAi mechanism for symbiotic stability interesting (if it holds up in other organisms) is that it could develop without requiring thousands of years or more of coevolution between a host and a symbiont. It should work for any endosymbiosis, as long as there is enough overlap of genetic sequences between the two organisms, and as long as the endosymbiont’s genetic material isn’t fully digested and the host has an RNAi system.

Understanding the mechanisms that stabilize symbiotic relationships could provide valuable insights into the first eukaryotes, which seem to have evolved from ancient endosymbiosis. For any complex life to develop from an endosymbiotic partnership, there must be a period of initial stability upon which selection can act to drive its emergence, Jenkins said. Although simple prokaryotic cells (bacteria and archaea) don’t have an RNAi system identical to that of eukaryotes, the general principle demonstrated in the new work might still be relevant to eukaryogenesis: RNAi evolved very early in eukaryotic history, and some of its components may have been derived from ones in prokaryotes.

Nick Lane, an evolutionary biochemist at University College London who studies the origin and evolution of eukaryotes, noted that most endosymbioses end in failure, with one of the parties driving the other to extinction. “Failure is the rule, and occasionally it’s successful,” he said. “This is a very nice example of the kind of mechanism that can allow it to be successful.”

From this initial work, however, which still needs to be peer reviewed, it is difficult to discern whether the stabilization mechanism might just be an artifact of the researchers’ experimental protocol. William Martin, an evolutionary biologist at Heinrich Heine University in Germany who was not involved in the study, points out that when paramecia consume their algae in the dark, the digestion process is slow and partial; the antibiotic that Jenkins and his colleagues used to eliminate the algae acted fast. It’s possible that the reported RNAi interactions might not happen at the slower rate.

If future studies confirm these findings, this mechanism might have significance beyond the context of endosymbiosis. Jenkins said he hopes to use the results of this study to see if cells have similar built-in mechanisms to discourage cannibalism. If a paramecium ate another paramecium, for example, Jenkins said he would expect the floating genetic bits from the digested cell to also activate the aggressor’s RNAi mechanism, given their nearly identical genetic sequences.

“You can envisage this cost to cannibalism, or at least a cost to eating a very close relative, emerging through exactly the same mechanisms,” Jenkins said.