“In our efforts to streamline farming practices to produce more meat for more people, we have inadvertently created conditions by which a harmless parasite of wild ducks can be converted into a lethal killer of humans.”

-Johns Hopkins neurovirologist R.H. Yolken and Stanley Medical Research Institute director E.F. Torrey¹⁵⁶¹

Earl Brown, PhD., Professor, Biochemistry, Microbiology and Immunology, University of Ottawa

All bird flu viruses seem to start out harmless, arising out of the perpetual, benign, stable reservoir of innocuous waterfowl influenza. They start out as mild, low-grade, so-called LPAI viruses, low-pathogenicity avian influenza. H5 and H7 viruses, however, have the potential to mutate into virulent, high-grade, “fowl plague” viruses, now known as HPAI—highly pathogenic avian influenza.

HPAI viruses aren’t born; they’re made. The World Health Organization explains in its 2005 assessment of the pandemic threat: Highly pathogenic viruses have no natural reservoir. Instead, they emerge by mutation when a virus, carried in its mild form by a wild bird, is introduced to poultry. Once in poultry, the previously stable virus begins to evolve rapidly, and can mutate, over an unpredictable period of time, into a highly lethal version of the same initially mild strain.¹⁵⁶² Scientists have demonstrated this transformation in a laboratory setting. A collaboration of U.S. and Japanese researchers started with a harmless virus isolated from waterfowl, H5N3 from a whistling swan in this case, and proceeded to do serial passages through baby chickens. First, the researchers took day-old baby chicks and squirted a million infectious doses into their lungs. Over the next few days, the virus would presumably start to adapt to the chicks’ respiratory tracts, with the viral mutant that learned best, through trial and error, to undermine the hatchlings’ defenses selected to predominate. After three days they killed the chicks, ground up their lungs, and squirted the viral lung slurry down the throats of other chicks. They again allowed a few days for the virus to adapt further before repeating the cycle two dozen times.

When they killed the last set of chicks, the researchers ground up their brains instead of their lungs and infected five additional rounds of healthy chicks with the brain pulp. With every passage, the virus grew more adept at overwhelming and outwitting the fledging birds’ immune systems to best survive and thrive in its new environment. The final infected brain sample, after two dozen cycles though lungs and five cycles through brain, was squirted into the nostrils of healthy adult chickens. If you do this with the original swan virus nothing happens, but influenza is a fast learner. By the 18th lung passage, the virus was able to kill half of the chickens exposed. After the final five brain passages, the virus was capable of rapidly killing every single one. The researchers concluded: “These findings demonstrate that the avirulent [harmless] avian influenza viruses can become pathogenic during repeated passaging in chickens.”¹⁵⁶³

If mad scientists wanted to create a bird flu virus of unprecedented ferocity, they could try to continually keep cycling the virus through chickens. Imagine if the serial passaging was done not two dozen times, but 20,000 times. What kind of virus would come out the other end?

The World Organization for Animal Health and the Food and Agriculture Organization of the United Nations agree that it has been “prove[n]”¹⁵⁶⁴ that once LPAI viruses gain access to poultry facilities, they can “progressively gain pathogenicity in domestic birds through a series of infection cycles until they become HPAI.”¹⁵⁶⁵ However, deadly bird flu viruses don’t tend to arise in just any poultry operation. According to USDA researchers, it’s the “high density confinement rearing methods” that give bird flu “a unique chance to adapt to the new species.”¹⁵⁶⁶ That is, today’s intensive farming practices may remove the natural obstacles to transmission that prevent the virus from becoming too dangerous.

David Swayne is the USDA’s leading bird flu researcher. Director of the USDA’s chief poultry research laboratory, Swayne has authored more than 100 scientific publications on avian influenza.¹⁵⁶⁷ According to Swayne, there has never been a recorded emergence of an HPAI virus in any backyard flock or free-range poultry operation. This is not surprising.

Imagine an outdoor setting. A duck flying overhead drops a dropping laden with relatively innocuous virus into a grassy field through which a flock of hens is pecking. The hens may be exposed to the virus, but coming straight from waterfowl, the virus is so finely tuned to duck physiology that it may not gain a foothold before being wiped out by a healthy chicken’s immune system. That’s why in the lab, researchers injected infected lung tissue from one bird to another to facilitate transmission. “The conditions under which we generated highly virulent viruses from an avirulent strain are generally not duplicated in nature,” the research team admitted. “However, viruses with low pathogenicity can cause viremia in physically compromised chickens.”¹⁵⁶⁸ Viremia means successful invasion of the bloodstream by the virus, an incursion they deem more likely to occur in compromised hosts.

If an outdoor flock does manage to get infected, the virus still has to keep spreading to remain in existence. Influenza virus is rapidly killed by sunlight and tends to be dehydrated to death in the breeze. Its ability to spread efficiently from one chicken to the next outside in the open air is relatively limited. In a sparsely populated outdoor setting, there may simply be too few susceptible hosts nearby to passage between in order to build up enough adaptive mutations to do more than ruffle a few feathers. There was a deadly outbreak among wild sea-birds in South Africa in 1961¹⁵⁶⁹ and a 2004 outbreak on two ostrich feedlots,¹⁵⁷⁰ and rare sporadic outbreaks of highly pathogenic bird flu viruses date back more than a century,¹⁵⁷¹ but these seem to be exceptions to the rule. According to bird flu expert Dennis Alexander of the U.K.’s Central Veterinary Laboratory, with the possible exception of the ostriches (which were kept at unnaturally high stocking densities in reportedly stressful unhygienic conditions),³¹⁷² highly pathogenic influenza viruses have “never known to arise in an outdoor flock.”¹⁵⁷²

Now imagine the mad scientist scenario. Tens of thousands of chickens crammed into a filthy, football field-sized broiler shed, left to lie beak-to-beak in their own waste. The air is choked with moist fecal dust and ammonia, which irritates the birds’ respiratory passages, further increasing susceptibility in chickens already compromised by the stress of confinement. Since the birds are standing in their own excrement, the virus need not even develop true airborne transmission via nasal or respiratory secretions. Rather, the virus has an opportunity to be excreted in the feces and then inhaled or swallowed by the thousands of other birds confined in the shed, allowing the virus to rapidly and repeatedly circulate. With so many birds in which to readily mutate, low-virulence strains can sometimes turn into deadly ones. The dose of virus transmitted from one bird to another might also play a crucial role.

Earl Brown specializes in influenza evolution. He studies how the virus acquires virulence and adapts to new species. In a landmark study published in the Proceedings of the National Academy of Sciences of the United States of America, he showed “that a group of 11 mutations can convert an avirulent virus to a virulent variant that can kill at a minimal dose.” Instead of studying waterfowl viruses in chickens, he studied human viruses in a similarly unnatural host—mice. Mice can be experimentally infected with human flu strains, but they don’t get sick. The virus is adapted to taking over and killing human cells, not mouse cells. Human viruses tend to replicate so poorly within a mouse that the mouse’s immune system wipes it out before the virus can cause so much as the sniffles. If the human virus is dripped directly into a mouse’s nostrils, though, and then you grind up the mouse’s lungs and drip the viral-infected lung tissue into another mouse’s nose and the cycle is repeated, in as few as a dozen cycles, the virus, which started out totally harmless, comes out the other end causing fatal pneumonia in the mice.¹⁵⁷³

Investigators have found that the key to this ramping up of virulence may be the size of the infectious dose that gets transferred from one animal to another. In the swarm of viral mutants in every infected mouse’s lungs, odds are that there are few capable of “heightened exploitation of the host.”¹⁵⁷⁴ But, if these mutants never find their way into another mouse, they will be wiped out along with the others. So even if an infected mouse were to rub noses with another mouse and transfer some small level of infection, chances are slim that the rare virulent mutants would happen to find their way into the next mouse. In fact, since influenza is so sloppy at replicating, the majority of mutants may be relatively dysfunctional, so a small transferred dose may actually cause the virus to lose potency.¹⁵⁷⁵ So it’s not just the number of animals that doses of virus get transferred between—size really does matter. The greater the number of viruses transferred or “passaged” from one animal to the other, the greater the pool natural selection has to select from.

Professor Brown described it to me like trying to win the lottery. If you only buy a few tickets at a time, the chance of winning the jackpot is slim. But what if you bought all the tickets? The evolution of hypervirulence, as seen in viruses like H5N1, may require an enormous viral load seen only in rather unique circumstances—artificial inoculation in a laboratory or animals intensively confined in their own waste. Inside an industrial broiler shed there can be high dose transmission through thousands of hosts—what Brown calls the “nub and the crux” of the development of extreme viral virulence.¹⁵⁷⁶ Highly pathogenic bird flu viruses are primarily the products of factory farming.¹⁵⁷⁷