Read our COVID-19 research and news.

The Rapid Rise (and Fall?) of Controversial Theory That Lab Accident Caused Flu Outbreak

A retired plant virologist from Australia has caused an international ruckus by proposing that a vaccine-manufacturing accident may have created the virus driving the current swine flu outbreak. Many leading influenza researchers immediately dismissed the theory out of hand or flat-out refused to speak about it. “Oh, come on, forget about this,” said virologist Peter Palese of Mount Sinai Medical Center in New York City. But influenza scientists at the World Health Organization (WHO) and the U.S. Centers for Disease Control and Prevention (CDC)—who similarly have deep skepticism about the thesis—have been forced to examine the evidence and publicly discuss what to many is a far-fetched idea.

Adrian Gibbs, who worked at the Australian National University in Canberra for 39 years before hanging up his lab coat in 2005, came up with this hypothesis, and although his specialty is potato viruses, he’s no stranger to flu genetics or controversy: He earlier published studies in Science and Nature questioning the supposed bird origin of the strain behind the 1918 influenza epidemic.

As Gibbs explained to ScienceInsider, the latest outbreak led him to scour the public databases and compare the genes of the recently discovered A (H1N1) virus—which is a mix of swine, avian, and human influenzas—with its closest ancestors. He confirmed, as others have reported, that six of the eight genes appear to have come from North America and two others came from Eurasia, which suggested to him that there were two parental viruses that had “reassorted” (fluspeak for combined).

Gibbs, who still lives in Canberra, then took this seemingly innocuous insight into a realm that made the eyes of influenza experts go wide. Gibbs says he discovered that the virus’s eight genes had evolved at a much more rapid pace than expected. Basically, he calculated this by looking at the known mutation rates of influenza viruses and comparing the changes in the new strain with the changes in its closest relatives. “It’s clear that all eight genes have speeded up for the last 7 years,” says Gibbs. “That struck me as really, really strange.” He could imagine how one parental strain might do this. “But how in the heck do two viruses do it?” he asks.

Perusing Google, Gibbs found a possible explanation. “I bumped into an advertisement for an influenza vaccine for pigs that contained three different viruses,” he says. The vaccine was supposed to contain killed versions of the virus, but what if there had been a lab accident and some had survived? Specifically, both human and pig vaccine manufacturers use eggs to grow the viruses, and Gibbs notes that studies have shown that this odd avian environment can accelerate their evolution. If these viruses were not properly killed, they could have reassorted in a pig and created the new H1N1 strain. “Not killing off the virus could explain the whole thing very neatly, but it’s only one of several possibilities,” Gibbs says.

When Gibbs phoned someone whom he would only describe as a WHO friend who was “in the thick of it,” he says there was silence on the other end of the line. The friend agreed to read and circulate a paper Gibbs had written, which Gibbs sent on 8 May. The story soon leaked to the press, and the paper quickly made its way to others leading the effort against the swine flu outbreak, including the chief of CDC’s influenza division, Nancy Cox. “The premise that the rate of evolution of the individual genes in this virus is faster than normal is wrong,” says Cox, who has done many studies of influenza evolution. Cox says that Gibbs chose the wrong old viruses to calculate how much change had occurred. In effect, he had improperly calibrated his clock, she says. “He gets an artificially faster rate of evolution,” says Cox. “So there’s no need to postulate an alternative host, whether it be eggs or the other natural hosts.”

Cox stresses that CDC has an open mind about possible origins of the virus, and as part of its own routine analysis had looked for evidence of lab contamination or error. “We had done all of those basic analyses very early on and had no indication that there was an unnatural origin for these viruses,” says Cox. And she notes that CDC puts the viral sequences in public databases so that investigators such as Gibbs can do alternative analyses. “This is what scientific inquiry is all about,” she says. “And when someone indicates that the origin might not have been natural, it’s important for us to go back and look at that methodology to understand why this research got the results it got.”

The bottom line is that Cox, Palese, and others say there’s no need to invoke a complicated theory for H1N1’s origin. “Our level of understanding of influenza viruses and the way they evolve and reassort in nature is such that postulating this is an enormous stretch, because we know that Mother Nature is so creative and provides so many different gene reassortants,” says Cox.

Gibbs says he doesn’t mind the criticism of his idea, and he has submitted his paper to an open-access journal so it can receive a proper airing. “Scientists are trained to be iconoclasts, not to believe what they’re told,” he says. “And they don’t believe until they’ve really had a good go at it.”