How Worms Avoid Eating Bad Bacteria and Warn Their Offspring Too
How Worms Avoid Eating Bad Bacteria and Warn Their Offspring Too

How Worms Avoid Eating Bad Bacteria and Warn Their Offspring Too

A small RNA in Pseudomonas triggers an avoidance response in C. elegans that can be passed on to the next generation, according to research presented at this week’s meeting of the American Society for Cell Biology.

Jef Akst
Jef Akst
Dec 12, 2019


This past June, Princeton University molecular geneticist Coleen Murphy and colleagues published their research documenting that after consuming a pathogen, C. elegans can pass on information about it to their offspring, allowing the next generation to avoid making the same mistake. But only some pathogenic bacteria trigger this transgenerational avoidance response. Murphy wanted to know why.

See “Worm Parents Pass on Behaviors Epigenetically to Offspring

Her group started exposing worms to various bits of pathogenic Pseudomonas bacteria, which the team had previously found to trigger the avoidance response across generations. To the researchers’ surprise, exposure to bacterial metabolites did not trigger an avoidance response, nor did bacterial DNA. Small RNAs in the bacteria, however, did. When they squirted a bunch of Pseudomonas small RNAs onto the worms’ usual diet of E. coli, the nematodes later avoided eating Pseudomonas, even though they’d never encountered the actual organism before.

The team looked for differences in the expression of small RNAs between Pseudomonas bacteria cultured at 25 °C—a temperature at which the microbes are pathogenic and trigger the avoidance response in C. elegans that consume them—and those cultured at 15 °C—conditions that result in no response in the worms—and identified six bacterial small RNAs that were upregulated in the bacteria kept at the warmer temperature. Further experiments with E. coli genetically engineered to express each of these small RNAs narrowed the search to one particular culprit that appears to trigger the avoidance response—even if the worms don’t actually get sick. “It’s like a false memory,” says Murphy, who presented the findings on Monday (December 9) at the American Society for Cell Biology annual meeting in Washington, DC, and earlier this year as a preprint on bioRxiv.

Small RNAs isolated from Serratia marcesans, a pathogen that does not trigger a transgenerational avoidance response, did not have this effect, she tells The Scientist.

Digging into the phenomenon further, her team found that the molecular pathways underlying the worms’ initial avoidance of Pseudomonas and their ability to pass that information on to their offspring appear to be one and the same. In fact, the small RNA’s signal must go from the gut to the germline before it can reach the neurons that control the avoidance behavior. “We got this crazy result early on,” Murphy says, referring to the signal’s route of transmission through the body. “We thought it was a mistake we made.”

But it was no mistake. In the study published in June, the team had found that lots of tiny C. elegans RNAs known as Piwi-interacting RNAs (piRNAs) were expressed differently after exposure to pathogenic Pseudomonas, and that knocking out prg-1, which encodes a regulator of piRNAs, in the germline blocked the transgenerational response.

In this latest project, the researchers found that knocking out prg-1 in the germline also blocked the avoidance response in the mothers. “Every single thing that was required for transgenerational inheritance was also required for avoidance in the mother,” says Murphy. “It basically is this system-wide signaling that the worms use to interpret what they’re eating.”

The team confirmed the results with wild C. elegans and wild bacteria, to demonstrate that this was not simply an artifact of the laboratory environment. Although the exact mechanisms by which the Pseudomonas small RNA triggers avoidance behavior—not to mention the passing of that behavior on to the next generation—remain a bit of a black box, Murphy and her colleagues did determine that they do not involve the known components of pathways for processing microRNAs or viral RNAs. “It’s something that’s completely new,” she says. “[It] opens up a whole new set of questions.”

Jef Akst is managing editor of The Scientist. Email her at