The viruses, Jeremy Barr realized, were in the wrong place.

Barr and his colleagues at San Diego State University had grown a layer of gut cells in a dish, much like those that line the surface of our own twisting intestines. The cells formed such tight connections with each other that bacteria couldn’t sneak past them. Even a dye couldn’t get through. The layer was meant to be impermeable, until the team infused the water on one side of it with viruses called phages.

After a few hours, they found a few of these phages on the other side. The cells had absorbed them at one end, and shoved them out the other. “It took us a while to realize what we were seeing, but when we did, it was really exciting,” Barr says.

Barr believes that the same process happens in our bodies, frequently and relentlessly. If he’s right, it means that our guts are absorbing billions of viruses every day, sending a steady stream of them into our bloodstream and the rest of our organs.

That’s not something to worry about. Phages don’t infect human cells and they don’t cause disease. Their full name, bacteriophages, means “eaters of bacteria,” and as that suggests, they infect and destroy bacteria. In doing so, Barr says, they could act as part of our immune system. They anchor themselves in the layer of mucus that lines our gut. By infecting the bacteria that also thrive there, they keep these microbial populations in check, and could determine which species get to live in our bodies.

This relationship is likely an ancient one. Mucus is universal to animals from corals to fish to humans, and phages are universal to mucus. Perhaps this was how the very first animals defended themselves against infections. They developed mucus to concentrate phages that were plentiful in their environment, and the viruses in turn helped their hosts to control the microbial multitudes around them. It was a mutually beneficial relationship between animal and virus, and one that continues today.

But the latest experiments from Barr’s team, many of which were done by his colleague Sophie Nguyen, suggest that this relationship between animals and phages is even more intimate. The phages aren’t just sitting atop human gut cells, acting as bouncers. They are actually being trafficked through the cells themselves. The team even used powerful microscopes to confirm the presence of phages within the cells. “A cell is enormous compared to a phage,” says Barr. “It’s like finding a cup of coffee by sectioning a skyscraper.”

In the experiment, just 0.1 percent of the total phages made it through. But based on their rate of travel, and the staggering number of them in the average human gut, the team estimated that our gut cells absorb around 31 billion phages every day. “The percentage feels like it can’t be that important but when you turn that percentage into absolute numbers, it feels biologically relevant,” says Corinne Maurice from McGill University, who also studies phages and was not involved in this study.

The team only did experiments using lab-grown cells, but Barr says there’s good reason to think that the same viral journeys take place in living bodies. For over 70 years, scientists have been “finding phages in parts of the body where they shouldn’t be,” he says, including supposedly sterile organs like the lungs. Microbiologist René Dubos found hints of this in 1943, by injecting phages into the guts of mice and finding those same viruses in the rodents’ brains.

“Phages can be detected outside the gastrointestinal tract, but there hasn’t been any real proof of how they get there,” says Lori Holtz from Washington University School of Medicine in St Louis. Many scientists believed that they were just leaking through gaps between the cells, but Nguyen’s work suggests that they are actually going through the cells themselves. In her experiments, the phages could traverse cells that line the kidneys, lungs, liver, and even the brain. “That’s absolutely astonishing in my view,” says Barr. The brain is separated from other organs by the blood-brain barrier—one of the most tightly controlled borders in the body. It’s incredibly hard for scientists to get small molecules through it. And yet, phages seem to do so.

This isn’t an infection in any meaningful way. The phages aren’t hijacking human cells to make more copies of themselves, as viruses like influenza, Zika, or Ebola might. Instead, Barr thinks that the cells are in control. They’re actively engulfing phages, and shuttling them from one end to the other. Why?

For a start, this would suffuse our bodies with a sparse but continuous stream of phages, which might then protect our organs against wayward, opportunistic bacteria. But Barr speculates there are more unexpected purposes at work. By sensing and studying the phages they absorb, cells could fine-tune the production of the mucus that houses these viruses, or the chemicals that feed the microbes that the phages then infect. If the cells break down some of the absorbed phages, they could access and use the viruses’ genetic material. All of this is possible, and none of it is certain. Scientists are only starting to eavesdrop on the three-way conversations between bacteria, phages, and our own cells. “It’s a big unknown,” says Barr.

Maurice also notes that phages aren’t just a homogenous group. They are incredibly diverse in their own right, and they exist in large communities. “Do they compete with each other?” she wonders. “Could some of them facilitate the entrance of others into human cells? I have no idea.”

These discoveries could have potential medical implications. For a century, scientists have looked to phages as a way of curing bacterial diseases, without having to resort to antibiotics. Although phage therapy fell out of favor in Western countries, research continued to blossom in Eastern Europe and Russia. And in recent years, there have been some spectacular successes, in which patients were pulled back from death’s door by infusions of these viruses

Martha Clokie from the University of Leicester notes that several infectious bacteria, including those that cause tuberculosis and Lyme disease, can enter and infect human cells. “If we want to treat these diseases, having phages that can cross into human cells would be very useful,” she says. “This is a neglected research field.”