Caused by a mutation, cystic fibrosis (CF) isn't contagious, but one serious complication definitely is: infection with Mycobacterium abscessus, an obscure agent related to the microbe that causes tuberculosis. Between 5% and 10% of CF patients become infected, and that number is growing. The bacterium thrives in the excess of thick mucus that builds up in the airways of CF patients—sometimes with fatal results.
Until recently, scientists believed that patients picked up the microbe at random, from the soil or water—making infection a case of bad luck. But an analysis of hundreds of bacterial genomes from patients around the world, published in this week's issue of Science, tells a different story. It suggests that the bacterium has adapted to humans and that several dangerous strains are spreading from one CF treatment center to the next, from country to country, and even between continents in a silent epidemic.
The researchers have no good explanation for this unexpected mobility, and not everyone is convinced. But other CF experts say the study shows that hospitals need to do more to reduce the infection risk for their patients. "This has huge implications for CF center isolation and cleansing protocols," says Brian O'Sullivan of the Geisel School of Medicine at Dartmouth College.
In CF a defect in a gene for a transporter protein involved in mucus production affects many organs and tissues, but its most serious effects are often seen in the lungs. The mucus-filled lungs are prone to infections, which lead to inflammation, which leads to more mucus production, worsening the disease or even suffocating the patient. M. abscessus, rare in healthy people, is notoriously difficult to treat because it is resistant to most antibiotics, O'Sullivan says. "Even when it seems to be gone it can resurface months or years later."
We need to rethink infection control measures within CF centers.
The first evidence that CF patients don't always pick up the microbe at random came in 2013. Researchers discovered that several patients at Papworth Hospital in Cambridge, U.K., were infected with bacteria that were almost identical genetically—something very unlikely to happen if the infections occurred independently. "That told us that there was transmission in one hospital and that was worrying enough," says geneticist Julian Parkhill of the Wellcome Trust Sanger Institute in Hinxton, U.K., who is an author on both the 2013 paper and the new one. "But it didn't answer two things: How widespread was that transmission? And was it just Papworth or was it happening elsewhere?"
To find the answers, the researchers collected more than a thousand M. abscessus isolates from 517 CF patients in the United States, the United Kingdom, Denmark, Sweden, the Netherlands, and Australia. They found that the microbial genomes from some patients differed widely from one another, just as you'd expect with environmental infections. But more than three-quarters of the study's patients had strains that belonged to "clusters" with very similar genomes—even though some of the patients lived far apart. "There are three major clones," Parkhill says, "and several others that are emerging and spreading."
The researchers found that bacteria forming part of a cluster were more likely to be taken up by human cells and to survive in them than the microbes without close relatives; the cluster strains also caused more severe disease in mice. This suggests that the microbes have adapted to cause more severe disease in humans, Parkhill says.
But though it's easy to envision spread within a single hospital—for instance through contaminated surfaces or droplets lingering in the air—how long-distance spread could happen is a mystery. CF patients don't travel much between centers, and equipment doesn't circulate widely, says study author Andres Floto of Papworth Hospital. It also seems unlikely that the bacteria are carried by an animal or some unknown environmental vehicle; many bacterial genomes are so similar that the spread must have been rapid. "Our most likely explanation (although we have no proof of this yet) is that healthy humans might be acting as vectors of transcontinental spread," Floto wrote in an email.
Erik Böttger, a medical microbiologist at the University of Zurich in Switzerland, says the paper makes a convincing case for spread within hospitals, which he says is important. But more evidence is needed about long-distance spread, Böttger says. He points out that the researchers didn't collect M. abscessus from the environment; patients in different countries may become infected with very similar bacteria because those are the most common in the environment worldwide, he argues. Parkhill admits that the lack of environmental samples is a weakness, but he points out that the big differences seen among bacterial genomes from patients who didn't belong to a cluster suggest that M. abscessus genomes in the environment vary widely.
Regardless of whether the clones are spreading globally or just locally, "we need to rethink infection control measures within CF centers," Floto says. At Papworth Hospital, room-cleaning protocols have already been improved, and the ventilation system in a new CF center has been redesigned to completely change the air once every 4 minutes.