Many thousands of years ago, on a chilly African night, a group of people gather around a fire in a cave. Using the flames, they cook their food, fashion new weapons, and warm themselves. But where there’s fire, there’s also smoke, and the smoke is giving the huddled humans a wretched cough. And in their inflamed airways, a microbe that normally lives in the soil is taking hold, changing, evolving into something new.

This, according to Rebecca Chisholm and Mark Tanaka, biologists at the University of New South Wales, is a possible origin story for one of the oldest human diseases—tuberculosis.

Caused by a bacterium called Mycobacterium tuberculosis, the disease formerly known as consumption has been plaguing people for tens of thousands of years. The scars of its infection still linger on 9,000-year-old Israeli skeletons and 2,000-year-old Egyptian mummies. Today, it still kills around 1.5 million people each year—more than any other infectious disease. And all of this, say Chisholm and Tanaka, began in fire.  

Scientists used to think that tuberculosis jumped into humans from some other animal, as is the case for many other diseases, like HIV, malaria, and Ebola. But they probably got the direction of spillover backwards: Based on current evidence, it seems that tuberculosis was first a human disease, which we then gave to other animals. All the bacterial strains that cause tuberculosis in seals, cows, rodents, and more originated from the bacteria that infected us. So where did those microbes come from?

The environment, probably. M. tuberculosis comes from a lineage of microbes called mycobacteria, which live in soil and water. They’re  mostly harmless. Some can cause opportunistic infections when they get into people whose immune systems are weakened or whose airways have been inflamed. But such cases are rare, and the bacteria behind them rarely spread to new hosts. By contrast, M. tuberculosis is what’s called an obligate pathogen—it lives only to sicken. And it’s highly contagious, spreading through the coughs and sneezes of its sick hosts.  

So how does an environmental microbe that occasionally causes disease become a human microbe that only causes disease? “It has to evolve to become transmissible between people,” reasons Tanaka. “That can be a really slow process if it doesn’t infect people very often. So maybe there’s a factor that accelerates this process, that gives the bacterium multiple chances to evolve.”

“Fire is a pretty good candidate,” he says. For a start, it’s dirty. When inhaled, particles in smoke can cause respiratory diseases and prevent the immune system from clearing away accumulated microbes. That makes it easier for an environmental mycobacterium to infect humans. And since fire brings people together, often within poorly ventilated places, it also gives a newly infective mycobacterium the chance to find new hosts. All the right evolutionary pressures were there for pushing a harmless soil microbes into a transmissible cause of disease.

To test this idea, Chisholm and Tanaka simulated the evolution of an ancient mycobacterium. They showed that even if that microbe needs just two mutations to become mildly transmissible between people, it’s very unlikely to accrue those under normal circumstances. But add fire to the mix—or more precisely, the increased social contact and damaged lungs that fire would cause—and the odds shoot up. It certainly wouldn’t be the first time that we wielded a new technology with unintended consequences.

For now, this is just a hypothesis. But it’s “really interesting and thought-provoking”, says Caitlin Pepperell, from the University of Wisconsin, Madison, who studies the evolution of human disease. “It’s plausible because smoke inhalation is so damaging to the lung’s innate immune system—our first line of defense against tuberculosis. Perhaps the bacteria that breached this defense had an easier time of it from that point on. Smoke inhalation also increases coughing and could enhance TB transmission.”

The timings fit, too, adds Tanaka. Although there’s a lot of uncertainty around the first uses of both controlled fires and tuberculosis, it seems likely that the former preceded the latter.

But all of this hinges on the idea that M. tuberculosis evolved first in humans before moving into animals. It certainly looks that way for now, but the story may change as scientists learn more about this microbe and its relatives.

“There may well have been another African mammal that was a reservoir for TB before it made its jump to humans,” says Kirsten Bos, from the Max Planck Institute for the Science of Human History. So far, scientists have only studied M. tuberculosis and its relatives in six animal lineages, so there’s likely a lot more diversity out there. And if they find an animal group whose TB strains predate the known ones, “that might rock the boat a bit on the Pleistocene campfire theory,” says Bos.

“It’s hard to test the hypothesis directly because it happened hundreds of thousands of years ago, and we can’t go back in time,” Tanaka concedes. But they might be able to find more time capsules—remains of ancient humans and animals that contain DNA from ancient strains of TB.

That’s very challenging, and doubly so because they would be looking not just for ancient samples of M. tuberculosis, but also the microbes that it evolved from. “And we have no idea what the ancestor of M. tuberculosis looked like, and distinguishing this proto-TB bacterium from the myriad contaminant mycobacteria in the soil seems near impossible to me,” says Pepperell. “But it’s worth trying!”