Fear of FLU – pandemic influenza outbreaks

Fear of FLU – pandemic influenza outbreaks – Abstract

Patricia Gadsby

Flu pandemics can envelop the globe in a lethal embrace. In 1918, the death toll neared 30 million. Where do deadly flus come from? Hong Kong’s bout with bird flu offers a scary new answer. By Patricia Gadsby

There was a moment at the peak of Hong Kong’s bird flu crisis, around Christmas 1997, that Kennedy Shortridge will never forget. Shortridge, an Australian virologist who has called Hong Kong home for almost 27 years, was in the Cheung Sha Wan Wholesale Poultry Market with a team testing poultry for the avian influenza virus designated A (H5N1). The previous spring, the virus had killed thousands of chickens in farms bordering southern China before fizzling out. In May 1997 it killed its first human victim, a three-year-old boy whose preschool had kept little chicks and ducklings. Now, with winter setting in, this fluky virus was breaking out in humans again–4 cases were confirmed in November, 13 in December. More alarming still, a third of the victims were dying. Shortridge knew that the birds would have to go.

In all his years of influenza surveillance, Shortridge had never seen anything like this. A sick chicken isn’t usually hard to spot, but this chicken was standing in its wooden cage, looking perfectly normal. And then, very gently, the bird just tipped over, and Shortridge saw blood trickling from its cloaca, the opening at its rear end. “Do we have a chicken Ebola on our hands?” he wondered, using Ebola as shorthand for the bird’s unbelievably fast hemorrhagic death. “In this one market, chickens were literally dying before our eyes. Were we on the verge of seeing a change in the virus? Were we going to see an explosion of virus across the markets of Hong Kong?”

One could take comfort in the knowledge that as long as H5N1 remained an “unreconstructed” bird virus, its opportunities for infecting people–now and then, and one at a time–were limited. But what if H5N1 mutated into a form more at home in humans? Or what if it joined in unholy matrimony with a human flu virus and created viral offspring that spread easily–flying from person to person through coughs and sneezes? “This had never happened before in history,” says Shortridge. “It was terrifying.”

Throughout that Christmas holiday, Shortridge and his increasingly worried–and exhausted–colleagues at Queen Mary Hospital continued the testing. About one-fifth of the chickens sampled were infected with H5N1. On December 29, even before the tests were completed, the government grimly went ahead with the slaughter of all the chickens in the markets and farms, well over a million birds, and began the monumental cleanup of the shuttered stores. Chinese New Year came and went, celebrated without the traditional fresh poultry dishes. No more human H5N1 cases. Months went by. The crisis in Hong Kong began to seem–in some quarters at least–less like a close call than an overreaction. Robert Webster, a virologist at St. Jude Children’s Research Hospital in Memphis, completely disagrees. “If this virus had really adapted to humans,” he says, “half the world’s population could be dead by now. We’d be looking at the next pandemic.”

Our world has been swept by three influenza pandemics in this century. The most devastating by far was the so-called Spanish flu in 1918: virtually every person on Earth was infected, and an estimated 30 million died, many more than those killed in World War I. The Asian flu of 1957 killed about 70,000 Americans (figures for the rest of the world are not known), and the Hong Kong flu of 1968 killed about 36,000. As far as we know, these flu pandemics, and the epidemics that occur like aftershocks in between them, are caused by the influenza virus’s changeable nature: epidemics result from genetic “drift”–tiny mutations just large enough to let the virus slip by many people’s immune systems; pandemics, on the other hand, involve a seismic “shift”–gene substitutions so large they leave pretty much everyone defenseless. But a host of questions remain: Where do these pandemic viruses spring from? What makes them so fierce? Did we have a lucky escape in Hong Kong? Is it only a matter of time before we face a flu as deadly as the 1918 virus?

Flu viruses belong to one of three families, designated influenza A, B, and C. Pandemics appear to be a specialty of type A viruses (a group to which the Hong Kong chicken flu belongs). Like all flu viruses, they use single-stranded RNA as their genetic material, and they make sloppy mistakes when they copy themselves, so bit by bit the viruses change–they drift from their original form. The changes that matter most are in the spiky surface proteins they use to infect cells in the human respiratory tract, proteins called hemagglutinin (H) and neuraminidase (N).

But type A viruses are odd: their RNA genome comes in eight segments. If two different type A viruses infect the same cell at once, they can shuffle their gene segments like cards in a deck. This reassortment can produce viral subtypes with combinations of genetic material unlike any our immune systems are used to, with genes that code for completely new H or N proteins, and maybe other proteins as well. By the late 1960s, flu researchers were beginning to understand that reassortment was behind the sudden shifts that underlie pandemics. They were also beginning to accept that the donors of the new, dangerous genes were probably animals.

But which animals? It had long been known that domestic pigs get the flu: every fall since 1918, pigs in the United States have come down with classic swine flu, a gift of the virus H1N1, which is a relative of the 1918 human virus. (The flu classification system numbers flu strains by variations in their hemagglutinin and neuraminidase genes.) The Hong Kong flu of 1968, however, had a hemagglutinin related to one found not in pigs but in birds. In the wake of that pandemic, Webster became part of an international surveillance team that would seek out the world’s wild flu reservoirs. By analyzing flu viruses in animals, researchers hoped to trace the source of the genes that mingled with human flu viruses to create such terrifyingly deadly strains.

During the 1970s and early 1980s, Webster and his colleagues at St. Jude amassed an impressive amount of evidence implicating birds–beginning in their own backyard. Right at home in Memphis, Webster found flu viruses in ducks returning from Canada in the fall. In Canada, he found “about 25 percent of newly hatched ducks had influenza viruses.” More viruses turned up in geese and gulls and in the small shorebirds that travel each spring between South and North America. “In a nutshell,” Webster says, you can find them “wherever you look in migrating waterfowl.” While humans get the flu by breathing in viruses, waterbirds are infected by the fecal-oral route, They pour out huge quantities of virus in their feces. Surprisingly, the viruses are passed from bird to bird but cause no disease at all. “Among them, these birds harbor all 15 of the known subtypes of influenza A,” says Webster, “and none of them get sick.”

The diversity of viruses in these birds hints at a long, shared evolutionary history. Moreover, the genes of these bird viruses are at an apparent evolutionary standstill–so nicely adapted to their hosts that new mutations offer no advantages and thus are not perpetuated. “That alone suggests to me we’ve described the main reservoir of the virus,” says Webster. “That and the fact that the viruses cause no illness.” Viruses don’t gain much by killing their hosts, and when a virus has a very long history in a species, it tends to reach a compromise with the animal’s immune system and cause little disease.

The problems come when flu viruses get into domesticated birds like chickens and turkeys–species in which they are not so well adapted–and the evolutionary brakes are taken off. The viruses make their new hosts sick, the hosts’ immune systems respond, and the viruses mutate to avoid the immune systems. They may even kill their hosts.

In the early 1990s, for example, a virulent outbreak in Mexican chickens was traced to a flu strain very much like the one found harmlessly infecting shorebirds. Flu outbreaks have also caused huge losses among poultry farms in the United States, many of which are clustered along the flight path of migrating waterfowl. The problem can spread to cities too. In December 1997, in fact, while Hong Kong was dealing with H5N1, New York was temporarily shutting down and cleaning up its bird markets in response to an outbreak of another avian virus, H7N2. H7 viruses, like H5 viruses, are notorious for mutating into frightful fowl plagues. New York officials are quick to point out that this H7 flu has never been a threat to humans. But of course, until the Hong Kong incident last year, the same was said about H5 viruses.

“I’m more concerned than I would have been pre-Hong Kong,” says Edwin D. Kilbourne, a member of the NIH’s Pandemic Planning Committee. “Now we’ve seen what an avian virus can do.”

There had already been evidence that the 1957 and 1968 flu pandemics were caused by human viruses that had substituted avian genes–probably from waterfowl–for some of their own. Yet catching flu from pigs seemed more likely. While humans don’t carry receptors for bird flu on their cells, pigs do. The going theory was that flu viruses might reassort more easily, and become more compatible with humans, if they mixed in an intermediate host like a pig. Pigs on farms in Asia (where both the 1957 and 1968 pandemics originated) are often kept close to ducks and chickens. And pigs have receptors in their snouts not just for their own swine virus but for bird and human viruses as well. So, potentially, a pig could snort up a bird virus in infected droppings or water, inhale a human virus spread by a coughing farmer, and become a mixing vessel for the two. The viral progeny might then infect humans nearby.

What the Hong Kong outbreak has now demonstrated is that a direct hop from birds to humans is absolutely possible. Maybe pigs aren’t as crucial as was thought. On the other hand, the pig-as-mixing-vessel theory is seductive, and it too has been observed: in 1993 a human virus, H3N2, and a bird version of H1N1 mixed in European pigs and produced a virus that infected two children in the Netherlands. But according to Yoshihiro Kawaoka, a virologist at the University of Wisconsin in Madison, there’s no direct proof that pigs were involved when bird genes appeared in the human influenzas of 1957 and 1968.

So who is giving what to whom? Bird viruses may have figured in the last two pandemics, but the closest relative to the 1918 Spanish flu has always looked like classic swine–the porcine H1N1 virus that also first appeared in 1918 and that still circulates in pigs today. By some historical accounts, humans actually got sick first–implying that people gave the virus to pigs. That global killer of 80 years ago is obviously the flu we most want to understand. Unfortunately, it is also the hardest to get a handle on. No one even knew influenza was caused by viruses in 1918, so the cause was essentially lost from the start. By the 1930s, when researchers realized a virus was responsible, the original, deadly 1918 virus was long gone. Or at least we thought it was, until 1997, when researchers began resurrecting its genetic ghost.

Jeffery Taubenberger did not start out as a flu researcher. He’s chief of the division of molecular pathology at the Armed Forces Institute of Pathology in Washington, D.C. He and his collaborators work on coaxing genetic information out of tissues fixed in formaldehyde and preserved in little paraffin blocks. The institute has a long history–it was founded by Abraham Lincoln, who realized that more soldiers died of infectious diseases than of gunshot wounds, and its tissue archives date back to the Civil War. Since formal cataloging began in 1917, more than 3 million cases have been documented, “showing a quasi-Victorian passion for collecting,” as Taubenberger puts it.

“In an era of budget cutting,” he adds, “we wanted to showcase the repository, so we looked for a good project.” That’s how he got into the 1918 flu business.

The first step was to look for tissues of 1918 flu victims. Of the 70 initial samples studied, just two were positive for influenza RNA. “The first case was a 21-year-old Army private from Fort Jackson, South Carolina,” recalls Taubenberger. “Amazingly, he died on September 26, 1918, the very same day the second fellow died, at Camp Upton in New York. The second soldier had one of those very unusual 1918 pathologies. He died in three days of massive pulmonary edema–lungs completely filled with fluid.”

The Spanish flu washed over the world in two waves. During the spring wave, it was very infectious but not very virulent. By September, the beginning of the fall wave, the virus was killing people in droves–especially young adults. Taubenberger had two questions: Where did the virus come from, and why did it suddenly become so ferocious?

His initial study, published in March 1997, focused on pieces of the hemagglutinin gene from RNA found in the first victim, which looked pretty much like classic swine flu, H1N1. And it still looks like a mammalian gene now that the whole of it has been sequenced, says Taubenberger. But, he adds, “it’s the most avian-like of mammalian Hi sequences. So probably, ultimately, it did come from a bird. The question is in the timing–how long ago?”

A clue might come from past studies that examined flu antibodies in the blood of people who were alive around the turn of the century. Some elderly people may have had antibodies to an ancestral H1N1 as early as 1905. So perhaps the virus skulked around for years before the Spanish flu erupted, sometimes getting into people but not yet capable of spreading easily or quickly. “Maybe,” speculates Taubenberger, “a human and an avian virus reassorted, and then the avian-derived virus took some years to fully adapt to life in mammals.

“Were we starting to see something similar last year in Hong Kong?” he muses. “Maybe Hong Kong last year was what was happening in 1905. It’s worth thinking about.”

In one respect, though, the 1918 flu virus is definitely not like the Hong Kong bird flu. In the chicken viruses H5 and H7, a simple mutation in the hemagglutinin gene can change a mild-mannered virus into an almost uniformly fatal pathogen. That mutation, a little genetic stutter, allows the virus access to cells beyond its normal range–not just in the bird’s gut and respiratory tract but in its heart, kidneys, and brain. This worrisome mutation was one of the hallmarks of the avian virus that caused unusually severe illness or death in 18 people in Hong Kong. But Taubenberger found that the 1918 virus lacks this mutation.

This is not to say that mutations for virulence could not exist elsewhere–and Taubenberger has acquired another source of 1918 flu genes to investigate. In August 1997, Johan Hultin, a retired pathologist from San Francisco, quietly flew to Brevig, an Alaskan village nearly obliterated by influenza in 1918. With the villagers’ permission and help from four local boys, he opened the graves of flu victims buried in permafrost, hoping to take samples of their lung tissues. One of them was an obese young woman. “She was lying on her back, and on her left and right were skeletons, yet she was amazingly well preserved. I sat on an upside-down pail, amid the icy pond water and the muck and fragrance of the grave,” says Hultin, “and I thought, `Here’s where the virus will be found and shed light on the flu of 1918.'” The young woman has become Taubenberger’s key “to sequencing the whole viral genome,” he says. The cold, and an insulating layer of fat in her skin, helped preserve her lung tissue for 79 years.

Hultin’s and Taubenberger’s successes fanned interest in a much larger expedition to the polar north in August. Led by Kirsty Duncan, a Canadian geographer, the team traveled to Spitsbergen, an island off Norway, to exhume the bodies of six miners who died in October 1918. Ground-penetrating radar studies hinted that their graves might be below the permanent frost line. Though the odds of finding live virus in their tissue were almost nil, isolation suits were worn and biosafety precautions taken.

When the graves were found, however, they were just above the permafrost. “The GPR let us down a bit,” says Webster, who’s slated to do some of the laboratory analysis. “The good news is we got lots of samples. The bad news is the tissue was not frozen.” At best, researchers will be contending with maddeningly short, degraded pieces of RNA, requiring two years to get “to where Jeffery is now.” To put the work in perspective, says Taubenberger, had live virus been found in Spitsbergen, the whole virus genome could have been sequenced in a week or two. In weeks, the Spitsbergen team would have been staring down the genes of one of the greatest killers on Earth–perhaps even seeing the nature of the beast it came from.

Chicken is on the menu again in Hong Kong, and the flu to worry about these days is not avian but financial, the current Asian contagion. There’s no doubt in Shortridge’s mind, though, that the world had a very close call. “I think Hong Kong made the right decision and averted a pandemic,” he says. “And we did it largely by watching animals.”

When agricultural authorities picked up a chicken-killing virus in the spring, it was identified as an H5 virus and passed along to a high-security lab in the United States. When public health officials spotted a peculiar virus in a sick little boy, they dispatched it to Dutch researchers interested in oddball viruses. And when it became clear that the chicken virus was the virus making humans deathly ill, Hong Kong did what it had to: it shut its markets and gassed the birds–not just chickens but ducks and geese as well, the aquatic birds that may have passed the virus to them. (The closest match so far for the hemagglutinin of the Hong Kong chicken virus is one from a Chinese goose.) Enormous changes have been wrought in the traditional market system. Chickens brought in from southern China, for example, are tested twice, once on either side of the border. And no longer can chickens be mixed with live ducks and geese. In an almost unimaginable break with Chinese custom, aquatic birds are not only taken to a separate market but killed in advance and sold as dressed poultry.

To reduce the chances of animal viruses hopping into humans, it’s vital to know which viruses spell trouble, and where the viruses are. For decades flu researchers have popped strains of interest into their freezers. These isolates now serve as a reference library to identify viruses and work out their family trees–the Dutch team used tests based on viruses from Webster’s freezer to identify bird flu in the Hong Kong boy. Still, the Hong Kong experience makes it clear that current levels of surveillance are not good enough.

The ultimate reservoir of the Hong Kong chicken virus will probably be found among migrating waterbirds. “But you can’t kill all the ducks and aquatic birds of the world–that would be absurd,” says Webster. “It makes you realize that influenza is a noneradicable disease.” Antiviral drugs can work against flu, but our best protection against new viruses is vaccines (see “Beating Bird Flu,” page 86). The tried-and-true way to make a vaccine is still to use a closely matched but safe version of the virus you want to combat. Right now, though, there are no good matches for the bird flu virus.

“It’s one year after H5N1. And do we have a vaccine on the shelf in case we need it?” asks Webster testily. “No. And why not? Because we haven’t found a good surrogate virus with which to make a vaccine. And why is that? Because we haven’t done the necessary surveillance. We should be looking all over Southeast Asia.”

At an NIH meeting in September, virologists and public health officials discussed the need for more surveillance in nature. And vaccines for all 15 subtypes, it was decided, should be readied lest an H5 virus–or another new virus subtype–take us by surprise.

For now, researchers are catching their breath, immensely relieved that H5N1 was nipped in the bud. But Webster feels that another hop from birds to humans will occur in the near future. The Hong Kong outbreak, he adds, “served as a warning of what could happen. It was a killer–like 1918 on its way.”


Vaccines work by giving the immune system warning about a pathogen, and flu vaccines usually do so by introducing signature flu proteins into the bloodstream. But vaccines for a bird flu virus, like the one implicated in the Hong Kong outbreak, pose a problem. The H5N1 virus kills chickens, and fertilized chicken eggs infected with flu are the nurseries in which flu proteins for conventional flu vaccines are produced. (Eggs are used because the virus grows quickly in a chick embryo, and eggs are abundant and inexpensive compared with other production systems.) Still, there is hope. A couple of groups are working on vaccines against H5N1. And one such vaccine–developed by Protein Sciences of Meriden, Connecticut–has been used among workers studying the H5N1 virus in high-level biosafety labs.

The Protein Sciences method uses caterpillar cells as nurseries for a virus engineered to produce the H5 protein–the hemagglutinin variant that is thought to allow H5N1 access to cells throughout the human body. The protein can be produced from caterpillar cells far more easily than from eggs. A vaccine made in this fashion not only protects chickens against H5N1 but is also safe in humans. Whether it can defend humans against H5N1 can’t be known until someone is exposed to the bird flu–either by accident or outbreak.

Another group, Aviron, in Mountain View, California, has recently created a live-virus human flu vaccine (delivered via nasal spray) and has also experimented with a live-virus vaccine against the H5N1 strain. First they inserted the H5 gene from the Hong Kong flu virus into the weakened strain they use in the human vaccine. Then they grew the resulting virus in chicken eggs, like a conventional flu vaccine. When chickens are vaccinated with this weakened live virus, it primes immune cells in their airways to fend off the H5N1 strain. This vaccine protects chickens, but its benefit to humans remains unknown.

Other labs are hoping to concoct a vaccine from an avian flu virus that closely resembles H5N1 but doesn’t cause disease. The closest match so far is a flu strain found in the Singapore duck.

Both Protein Sciences and Aviron say their methods permit them to start making vaccines within two months of getting a new flu strain to work with. If we face an outbreak a la Hong Kong, and an appropriate new vaccine is on the shelf, it would be tested for safety first, then possibly distributed within three to four months. For now, researchers can only hope that the virus cannot move faster than that.

PATRICIA GADSBY (“Fear of Flu,” page 82) is a contributing editor of DISCOVER. “I was familiar with Southeast Asia’s `wet’ markets but was quite unaware of the live bird markets under my nose in New York,” she says. “There are over 70 in the metropolitan area, selling not only live birds but goats and other animals. I spent a sweltering morning changing in and out of paper jumpsuits, following two inspectors of the New York State Department of Agriculture and Markets who were swabbing chickens in the Bronx.”

COPYRIGHT 1999 Discover

COPYRIGHT 2000 Gale Group