Henipaviruses: Emerging Paramyxoviruses Associated with Fruit Bats

Researchers in Malaysia have confirmed a link between fruit bats and a recently discovered virus that killed more than 100 people last year and compelled authorities to slaughter roughly a million pigs.

Prof. Kenneth Lam Sai Kit, a microbiology professor at the University of Malaya, said his research team had successfully isolated strains of the deadly Nipah virus in urine collected from island flying foxes, one of the many species of fruit bats that live throughout Southeast Asia.

The finding confirms that the bats are carriers or reservoirs for the virus, which last year crossed over to pigs and then to humans. With no known vaccine or cure, the Centers for Disease Control and Prevention has put Nipah, which was named for one of the villages it hit, in the same class of viruses as ebola and AIDS.

Other scientists called Professor Lam's findings a breakthrough in understanding the virus and helping to prevent future outbreaks.

''It's very, very important that Professor Lam did find the virus,'' said Dr. John S. Mackenzie, a professor of microbiology and parasitology at Australia's University of Queensland and an expert on diseases transmitted from animals to humans. ''The final proof requires that you isolate the virus from the bats.''

Professor Lam's finding came just after an outbreak of Nipah in June in northern Malaysia. Health authorities in northern Perak state killed more than 1,700 hogs to prevent the disease from spreading after a woman living nearby was found to have the virus.

While it remains unclear exactly how the virus is transmitted from fruit bats to pigs, numerous opportunities exist for the rare crossover to occur, Professor Lam said. Fruit trees in which bats forage for food are often near pigsties, he said, and Malaysia's pig farmers tend to live very close to their pigs. Malaysia is considering measures to protect its $400-million-a-year pig farming industry, which was crippled by last year's outbreak.

Nipah first appeared as a pandemic early last year when it hit pig farms in Malaysia and slaughterhouses in nearby Singapore. Health workers initially believed they were dealing with an outbreak of Japanese encephalitis, which is transmitted from pigs to humans by mosquitoes. The symptoms were similar: high fever, aches and coma. Roughly 4 of every 10 sufferers died.

Confounding this early diagnosis was the fact that some victims had been inoculated against Japanese encephalitis and that most lived near or worked on pig farms. The virus was also killing some pigs, and Japanese encephalitis does not do that.

After a Malaysian microbiologist isolated the virus from the brain and spinal fluid of Nipah victims, scientists at the C.D.C. determined that it was not Japanese encephalitis but a new member of a class of viruses called paramyxoviruses that includes mumps and measles. Its closest relative, they found, was another virus called Hendra discovered in 1994 after killing 14 horses and their trainer in Australia.

Hendra was carried by fruit bats, so researchers began searching for a similar link in Nipah. Scientists from the C.D.C. and from Taiwan, Japan and Australia flew to Malaysia, scouring pig farms, villages and trapping bats. Tissue samples from roughly 300 fruit bats revealed antibodies against Nipah, ''but we were never able to isolate the virus per se from any of our samples,'' said Dr. Hume Field of the Queensland Department of Primary Industries, who spent six weeks in Malaysia looking for Nipah's source.

With no new evidence a year later, Professor Lam and his team set out to search themselves. In early June, he said, they spent three nights laying plastic sheets under trees full of flying foxes, collecting urine samples. Back in the lab, they found strains of Nipah that matched those recovered from humans.

Though only just discovered, Nipah appeared to be a very old virus, judging from how different it was in form, Dr. Field said. Like ebola and AIDS, he said, it may have infected people on a much smaller scale years before making headlines.

Researchers will now need to determine what environmental changes made Nipah so virulent. Professor Lam said that only three of his 1,000-odd samples contained the virus, which indicated that bats were by no means infected by the virus. And Professor Mackenzie, who researched the Hendra virus's transmission from bats to horses and to humans, said infections took place only under very rare circumstances, usually by direct contact with infected blood. In Nipah's case, no transmission between people appears to have occurred.

''What we've probably got is one spillover event where one flying fox that was excreting virus came into contact with an unfortunate pig that consumed either the discharge or the bat,'' Dr. Field said.

Professor Lam said that because Nipah can also be found in saliva, it was possible that pigs may have caught the virus by eating fruit that the bats had nibbled on.

How the virus moves from pigs to humans remains unclear, but researchers believe one possibility is that people breathe in infected particles of saliva coughed up by Nipah-infected pigs.

Professor Lam's findings also raised concern about the possibility that the bats, which are migratory, could carry the disease beyond Malaysia. Southeast Asia is home to several species of fruit bat, many of which are considered delicacies and have been hunted nearly to extinction. In Malaysia, for example, fruit bats are now protected by law.

Professor Lam said antibodies against Nipah had been found in other species of fruit bat. Moreover, Nipah is also known to infect dogs, cats and horses, presenting the possibility that pig farms may not be the only venues for crossover.

1. SOUNDBITE (English) Doctor [Dr. John S. MacKenzie (born 1943(est.))], Team leader and virologist from the University of Queensland in Australia:   "We hope it will be very productive from both sides. From the global perspective - in terms of knowing more about SARS - but also from the more local perspective, shall we say, in terms of us providing whatever information we can, whatever help we can to our colleagues here in China."

2. Pan from Dr Meirion R. Evans, UK epidemologist (left) and [Dr. Wolfgang Preiser (born 1965)], German virologist to Dr Robert F Breiman, US Infectious disease specialist (left) and MacKenzie

3. SOUNDBITE (English) [Dr. John S. MacKenzie (born 1943(est.))], Team leader and virologist from the University of Queensland in Australia:   "We've had 27 suspected cases in Australia but it's only information of the sort we're getting now - I mean, like we've been getting for the last three weeks - that we've been able to say that those 27 suspect cases in Australia are not SARS. And sure they've travelled in the right place. Sure they've got atypical pneumonia but they're not SARS under the current case-definition. So to be able to find out what the best definition possibly is, is incredibly important."

4. SOUNDBITE (English) Dr Robert F Breiman, American Epidemiologist and Infectious disease specialist:  "In the case of Legionnaires disease and (unclear) virus - and there have been many other emerging diseases - there is plenty of evidence that those diseases existed before. What's made it recognised is the actual clustering, it's the epidemic."

5. SOUNDBITE (English) Dr James H. Maguire, Chief of epidemiology, US National Centre for Infectious Diseases, Centres for Disease Control and Prevention:   "It may be, it may not be but it may be that what we learn, what is learned here in China will provide the key. There may be issues here. The fact that if this was really the first location for where these diseases occurred, there may be some reasons why that happened here that I think are gonna be very interesting to explore."

STORYLINE:   A team of World Health Organization (WHO) experts said on Monday they were analyzing the cases of Chinese sickened in an outbreak of a fatal flu-like illness in hopes of finding out whether it is linked to a mystery disease that has spread to three continents. The five-member team arrived on Sunday and said they would examine Chinese records this week.  They said they had not decided whether to visit the southern province of Guangdong, where five deaths from atypical pneumonia were reported. WHO says it is treating the outbreaks as part of the same emergency that has killed 10 people in Hong Kong, two in Vietnam and two in Canada, though it isn't clear if the deaths were caused by severe acute respiratory syndrome, or SARS. Another 305 people also were sickened in Guangdong since the first cases were reported in November, Chinese health authorities say. The Chinese government didn't report the outbreak until February and its reluctance to release information fueled a panic in Guangdong and neighboring Hong Kong.  People bought up supplies of antibiotics and vinegar for use as a disinfectant. Cases of SARS have also been confirmed in Europe.

It seemed like the typical viral ''summer cold'' as it began spreading through a nursing home in the Vancouver area in early July. Then in August, test results from the Canadian National Microbiology Laboratory suggested that the ''cold'' might be from a mutated SARS coronavirus.

Just the hint created deep concerns in a country that had already been hobbled by an outbreak in Toronto last spring. British Columbia officials temporarily imposed infection control and quarantine measures that had contained the Toronto outbreak.

The disruptive and costly actions turned out to be excessive. British Columbia officials have since determined that the SARS coronavirus did not cause the outbreak. Rather, they said, its virological cousin, the OC43 coronavirus, which commonly causes colds, was the probable culprit.

Disputes among Canadian officials over the virological findings have exposed deep flaws in the systems available to detect and monitor any potential outbreak in Canada and elsewhere. Not only did they underscore the need for more uniform laboratory testing, but also prompted health officials to renew calls for reliable tests to detect SARS, or severe acute respiratory syndrome, in its earliest stages of infection.

''The safety net for detecting the possible return of SARS is pretty fragile,'' said Dick Thompson, a spokesman for the World Health Organization, the United Nations agency in Geneva with responsibility for controlling the disease. ''It holds together, but it is spit and luck.''

Examination of the disputed findings in Canada is also occurring as the world is bracing for the possible resurfacing of SARS. A recent report from the United States National Intelligence Council -- which consists of experts from the academic and private sectors and which reports to the director of central intelligence, George J. Tenet -- lays out three bleak possible situations for the virus's re-emergence. Surprisingly, the intelligence report does not consider an instance in which the disease does not reappear.

The three situations outlined in the report are disconcerting. One is a resurgence in countries with major trade centers as international travelers spread SARS the way they did this year. A second is that SARS cases will occur sporadically but will be detected before the disease can spread, creating more of a public health nuisance than a crisis. The third unfolds in developing countries with weak public health systems.

All make early detection a priority. Canadian and United States health officials and SARS experts are expected to discuss what steps may reliably be taken to identify possible new SARS cases, among other topics, at a two-day closed meeting in Ottawa that begins tomorrow.

To some health officials, the outbreak at the Kinsmen Place Lodge nursing home in Surrey, British Columbia, has not only revealed important gaps that need to be addressed, but it may also provide lessons that can be used in the face of a recurrence of SARS.

''In a way, it was lucky that this outbreak happened before the influenza season because it is better to learn about the problems now than then,'' said Dr. Katrin Leitmeyer, a W.H.O. epidemiologist and virologist who went to the National Microbiology Laboratory in Winnipeg to review its findings.

One of the most glaring problems uncovered by the nursing home outbreak was the absence of a formal agreement among scientists about precisely what steps and laboratory methods should be used to make a definitive diagnosis of SARS.

For example, many test results remain in dispute, largely because the scientists involved in different laboratories did not use the same methods to test each specimen to try to identify the virus and to determine whether the patients' immune systems had produced antibodies against it. And partly because little is left of the specimens collected from the nursing home patients, it will take weeks more to prove that the suspected OC43 coronavirus, is the cause.

The conflicting laboratory results have led Canadian scientists and officials to clash over the validity and meaning of the tests performed on the lodge residents and staff members.

''We're at loggerheads with the National Microbiology Laboratory,'' said Dr. Mel Krajden, the chief virologist at the British Columbia Center for Disease Control.

The center's chief epidemiologist, Dr. David Patrick, said that from July 1 through August, 95 of the 142 lodge residents, or 67 percent, developed the summer cold and that 8 deaths were related to the outbreak. Of the 160 health workers there, 53, or 33 percent, developed the summer flu.

Initially, Dr. Krajden's team found no evidence of viruses like influenza, parainfluenza and respiratory syncytial virus that are common causes of upper respiratory illness.

SARS was low on the candidate list because the death rate among lodge residents was 8 percent, far lower than the rate of 50 percent or more found among people 60 and older in the SARS epidemic last spring.

But when a second wave struck the nursing home and the center still could not identify a virus, health officials reconsidered SARS. ''So, having come up empty-handed, we said why not send some of the very limited remaining materials to'' the national laboratory, Dr. Krajden said.

It was a step that many laboratories less concerned about the threat of SARS might not have taken under the same circumstances, health officials said.

The surprise came when [Francis Allan Plummer (born 1952)], the director of the National Microbiology Laboratory, said that his team had found tentative evidence of what might be a mutant SARS virus.

Then on Aug. 22, provincial officials said the latest findings from the British Columbia Center for Disease Control and the British Columbia Cancer Agency's Genome Sciences Center ''provide conclusive evidence'' that the virus responsible for the nursing home outbreak was not the SARS coronavirus, but probably its cousin, OC43.

''We have clearly found large sequences of the virus that are not present in the SARS coronavirus,'' Dr. Patrick said.

Last week, Dr. Larry J. Anderson, an expert on respiratory diseases at the Centers for Disease Control and Prevention in Atlanta, said his team had completed tests to detect antibodies to the SARS virus in the blood specimens sent by Dr. Plummer. The C.D.C. tests did not confirm the Canadian National Microbiology Laboratory findings that SARS antibodies were present.

Dr. Plummer said he agreed that the virus was not the classic SARS virus, but that his team had found no evidence of OC43.

''We do not dispute their results, but we also believe our results,'' Dr. Plummer said. ''Now what produced them is a puzzle.''

Among the possible explanations for the conflicting results is that two or more viruses were simultaneously affecting the nursing home. Another possibility is that the specimens from the lodge were inadvertently contaminated with SARS virus in the Winnipeg laboratory. Dr. Plummer said he considered that unlikely because of the pattern of his test findings. ''All these diagnostic tests are new and every laboratory has developed its own test,'' he said.

''We have a limited understanding of their performance, so comparing what one laboratory is talking about to another is problematic,'' Dr. Plummer said. ''This is obviously new territory for all of us. We are probably at a normal state of development for a disease that has been around for a few months, and we will be learning every time one of these things happen.''

The Canadian laboratories have not succeeded in growing the nursing home virus and then fully mapping its genome -- two findings that would provide the most definitive evidence of the causative agent. But virologists say that growing the coronavirus apparently involved in the nursing home outbreak is more difficult than growing the SARS virus in the laboratory.

The threat of SARS may prompt several changes in laboratory practices, virologists said. For example, diagnostic laboratories rarely include coronaviruses in screening tests for the cause of outbreaks of respiratory illness because they generally cause mild illness and because budgets preclude searching for every virus.

''No one ever cared that much,'' said Dr. Kathryn V. Holmes, a professor of microbiology at the University of Colorado Health Sciences Center in Denver who has studied coronaviruses for 20 years. She was not involved in the Canadian investigation.

And the nursing home developments raise questions about how research is conducted in a public health crisis, said Dr. Roland Guasparini, chief medical health officer of the Fraser Health Authority near Vancouver, which had jurisdiction over the nursing home outbreak.

[Dr. John S. MacKenzie (born 1943(est.))], an Australian virologist who is temporarily helping W.H.O. deal with the threat of SARS and other emerging diseases, said the nursing home episode pointed out three major worries about laboratory testing for SARS that an advisory panel of W.H.O. would need to address in October.

One is the lack of quality assurance. The various methods have not been tested in different laboratories under different conditions, Dr. MacKenzie said, and ''this is a big concern.''

Second, there is no internationally accepted standardization of the biological materials known as reagents used in testing in different laboratories. Dr. MacKenzie said he was particularly concerned about the lack of standardization of the control tests, which play a routine part in laboratory testing.

Third, Dr. MacKenzie said, ''we need to look very carefully at what we do if we have a positive case.''

''What's crucial is that a second laboratory, preferably an international reference laboratory outside the country that found the positive, confirms the tests,'' Dr. MacKenzie said. ''Some countries are not keen on this,'' he added, yet ''it is a major issue.''

At the October meeting, he said, ''We're going to go into detail about what is known, what isn't known, what needs to be done, and then prioritize some things that need to be looked at.''

But because the W.H.O. meeting will not be held until late October, it may be too late to be effective, Mr. Thompson, the agency spokesman, acknowledged.

By then the Northern Hemisphere may be experiencing the usual seasonal outbreaks of influenza and other respiratory illnesses. Under such circumstances, the continuing lack of a reliable diagnostic test for SARS could create chaos from the continuing inability to distinguish SARS from other illnesses that, by coincidence, were producing similar symptoms like fever, headache and cough.

An expert panel convened by the World Health Organization called on all countries yesterday to stringently oversee laboratory safeguards, to prevent another escape of the SARS virus like the incident that infected a research worker in Singapore in August.

If laboratory safety standards are not strengthened, ''the whole world can be vulnerable'' to another SARS epidemic, said Dr. John MacKenzie, an Australian microbiologist who attended the meeting in Geneva.

All countries should conduct inventories at their laboratories to determine where the SARS virus is being kept and what strains of the virus are being stored, Dr. MacKenzie said. Also, Dr. MacKenzie said, the World Health Organization, an agency of the United Nations, and member countries need to develop procedures to control which laboratories can hold and work with the virus, which at present apparently only exists in laboratories.

Angus Nicoll, director of communicable disease surveillance for Britain's health protection agency and the meeting chairman, said the committee members asked whether all laboratories were meeting the correct safety standards. ''The answer is probably not,'' Dr. Nicoll said in a telephone news conference.

The W.H.O. invited 45 international experts in public health, SARS, virologists, social sciences, laboratory science and other disciplines to meet for the first time to identify the most important gaps in knowledge about SARS, or severe acute respiratory syndrome.

The committee came up with a list of the highest priority research questions for researchers, but did not specify what the issues were. In the case this summer, a 27-year-old doctoral student accidentally became infected with SARS while working on the West Nile virus in a laboratory in Singapore. The student did not know he had been exposed to the SARS virus.

A W.H.O. committee that investigated the Singapore case identified flaws in laboratory procedures and concluded that the student became infected in the laboratory. The committee found that the West Nile virus had been contaminated with the SARS virus.

Besides laboratory safety, another priority on the committee's list is determining which animals, if any, are the sources in nature of the human SARS virus.

Although scientists in Hong Kong and China have found the SARS virus in exotic animals sold in markets in Guangdong Province, they have not determined whether the animals were infected in the wild or contaminated in the market.

''One of the very important research priorities will be to identify the animal reservoir, if there is one, from China,'' said Dr. Joseph Sung of the Chinese University of Hong Kong and Prince of Wales Hospital there.

Scientists have found evidence of SARS infection among animal handlers in the market. But, Dr. Nicoll, the British expert, said, ''we don't know quite what it is that they're doing that puts them particularly at risk.''

W.H.O. has held many meetings on SARS, but the two-day meeting that ended yesterday was the first to include social scientists. They recommended that researchers now try to determine the common misperceptions, myths and institutional hurdles that caused fear among the public and impeded control of SARS.

Health officials need to learn good ways to avoid confusion in informing the public about what is known and unknown about SARS, because there were many myths about how an individual could and could not catch SARS, participants said.

Social scientists also recommended additional research to help health officials deal with Chinese and other groups that were vulnerable to stigmatization in the SARS epidemic.

Dr. Jong Wook Lee, the director general of the World Health Organization, assured the committee that the agency would devote more money for SARS, if needed. But, Dr. Nicoll said, ''we are wary'' about the the agency's ability to deal with SARS and other epidemics of new and old diseases.

''We have learned a lot in the past year, but one of the things we have learned particularly is that we very much need W.H.O. for this kind of emergency,'' Dr. Nicoll said.

An independent scientific advisory panel convened by the World Health Organization yesterday recommended against routine testing for the SARS virus unless a cluster of cases develops and all other infectious agents have been ruled out.

The panel plans to soon publish a guideline for doctors and laboratories to follow to determine when to test for the SARS virus. Except for a laboratory worker who became infected, no case of the disease has been detected since the early summer when the W.H.O., a United Nations agency in Geneva, said the epidemic was over and transmission had stopped .

If a laboratory does come up with a positive test result for SARS, it must send the specimens to a second laboratory that belongs to the W.H.O. network for independent validation of the findings, said Dr. John MacKenzie, an Australian microbiologist who organized the science panel's meeting in Geneva this week.

Unless there is such confirmation, the United Nations agency will not list any country as SARS-infected, Dr. MacKenzie said in a telephone interview from Geneva.

If a major outbreak of SARS, or severe acute respiratory syndrome, occurred, then the system of sending specimens for independent confirmation would become less critical, said Dr. Malik Peiris of the Queen Mary Hospital in Hong Kong. W.H.O. credits Dr. Peiris with discovering the SARS coronavirus.

W.H.O. said it recognized the risk that SARS might spread if tests were delayed. ''We admit that we might well miss the first case or even the first cluster,'' Dr. MacKenzie said.

The panel, he said, is recommending limiting SARS testing for now largely because scientists do not have enough specimens of the serum portion of blood from patients who were infected during the epidemic earlier this year. The serum contains antibodies to the virus and is needed for purposes of scientific controls in diagnostic testing.

Laboratory scientists ''need positive serum to make quite sure that what you're seeing is what you expect to see,'' Dr. MacKenzie said.

But, Dr. MacKenzie said, the agency also is trying to prevent other risks, like a flood of unnecessary testing and false alarms.

''No lab test is perfect,'' he said, and the lack of serums for testing ''is a major problem.''

False positive test findings are more likely to occur when a disease is not occurring, and they would set off immediate time consuming, cumbersome and costly public health action. That includes isolating patients, tracing their contacts and imposing quarantines.

The panel's concern grows out of a situation in Canada this summer when the national laboratory initially said it had detected the SARS virus among residents of a nursing home near Vancouver. But a SARS laboratory at the United States Centers for Disease Control and Prevention in Atlanta did not confirm the findings.

Dr. Peiris said that the panel was now recommending a number of steps to double-check the first positive result, at the initial laboratory. The steps include conducting a different test, using another specimen from the same patient and retesting the original specimen.

Serum from infected patients is in short supply because the amount of blood that doctors remove for SARS tests is far smaller than what is needed for supplying laboratories around the world. To generate a larger supply, scientists in a number of countries are recruiting patients who have large amounts of SARS antibodies, Dr. Peiris said.

During the epidemic, some individuals whose serums were rich with SARS antibodies had donated blood as an experimental treatment known as passive immunization for very ill SARS patients. Such donations are relatively easy to obtain because they are altruistic. But, Dr. Peiris said, ''when you talk about donating blood for diagnostic tests it seems a bit more remote.''

SARS is one of many viruses that can cause a lung condition known as atypical pneumonia, and many doctors would be likely to suspect SARS because of the epidemic earlier this year. But, Dr. MacKenzie said, ''we don't want to test everyone who has atypical pneumonia unless there is a cluster of cases for which there is no other alternative diagnosis and for which antibiotics do not work.''

A possible exception is when staff members in intensive care units develop atypical pneumonia. A large number of health care workers developed SARS in the initial phases of the epidemic earlier this year.

In a drab high-rise in Washington Heights, behind locked doors to which only four people have electronic keys, a scientist in a head-to-toe body suit transferred the SARS virus from a test tube to a dish, with a calm born of day-to-day familiarity.

A year ago, the Greene Laboratory of Columbia University's Mailman School of Public Health did not exist; today, the stuff of nightmares and headlines resides there -- SARS, West Nile virus, Bornavirus, anthrax. With remarkable speed, the lab has helped turn Mailman into a major player in research on the infectious disease outbreaks and bioterror threats that have been the focus of the world's attention in the last few years.

The school is also doing front-line work on chronic diseases like juvenile diabetes. If Mailman still cannot match the public health schools at Johns Hopkins or Harvard for resources and technical prowess, it is at least no longer playing in a different league.

''They've always been important, but their standing has been increasing and they've acquired enormous talent and resources,'' said [Dr. John S. MacKenzie (born 1943(est.))], an Australian microbiologist and scientific consultant to the World Health Organization, who led the first Western team to investigate SARS in China. ''They've become a leader in the field of early disease detection and rapid diagnosis.''

The elevation of the Mailman School's profile stems in part from a series of world events that have put far more emphasis on biological threats, natural or manufactured.

But it also results from years of effort by Mailman's dean, Dr. Allan Rosenfield, to increase the school's resources and change its focus. He said its budget has grown to $140 million from about $10 million when he took over in the mid-1980's.

Dr. Rosenfield decided to put more emphasis on high-level research, bioterrorism and disease outbreaks. ''The anthrax attacks were a wake-up call,'' he said. ''We stressed the laboratory systems in this country -- several cities, a few states, even the C.D.C. -- beyond their capabilities. The ability to respond to these crises was clearly wanting.''

This spring, scientists at the new Jerome L. and Dawn Greene Infectious Disease Laboratory, headed by [Dr. Walter Ian Lipkin (born 1952)], developed a test for SARS, and Dr. Lipkin flew to China to donate 10,000 SARS test kits. He quickly established that test as one of a handful that are widely used for the virus, and himself as an adviser to the Chinese government.

Scientists at Mailman are studying, among other things, the West Nile genome to learn why it spreads faster than related viruses, evidence that SARS can hide in people's intestines without making them ill, whether an enterovirus might trigger juvenile diabetes, possible links between Bornavirus and psychiatric disorders, and the extent of the nation's preparedness for a terrorist attack.

In April, the Mailman school created a National Center for Disaster Preparedness and hired Dr. Irwin Redlener, president of the Children's Health Fund and a prominent voice on bioterrorism and other issues, as its first director.

Dr. Rosenfield lured Dr. Lipkin from the University of California at Irvine, making him director of the Greene laboratory, and Dr. Lipkin, in turn, brought much of his research team with him and recruited other people to the lab, which has about 30 researchers. Dr. Lipkin was the first to identify a 1999 outbreak of encephalitis as the first appearance of West Nile virus in North America, and he helped develop tests for West Nile.

Dr. Lipkin said he was very reluctant to leave California, ''but I recognized that the resources here would allow me to do work I couldn't do anywhere else.''

A $6 million gift from the family of Jerome L. Greene, a philanthropist who died in 1999, created the laboratory and endowed the director's chair. The lab has an area designated as biosafety Level 3, or B.S.L. 3, where researchers can work with dangerous contagions that were off limits to Columbia before.

Anyone entering the B.S.L. 3 lab must wear a body suit, mask and gloves, and pass through an antechamber akin to an airlock; some also wear hoods attached to whirring battery packs, pumping filtered air to keep germs away from their mouth, nose and eyes. Ultraviolet lights kill stray bacteria. Anything taken out must either go through a high-pressure autoclave to be sterilized at 270 degrees Fahrenheit, or be swabbed down with germicide.

Air pressure inside the lab is kept low so that air, and microbes, can only flow in without escaping. It has a sophisticated air filtration system and backup system, and its own emergency power generator. Security in the building is tight; Dr. Rosenfield said not even he has access to the area that includes the B.S.L. 3 lab.

Not many universities have comparable labs. Even some state health departments do not.

Across the metropolitan region, universities, government agencies and companies banded together this year to form the Northeast Biodefense Center, headed by Columbia, the New York State Department of Health and the Academic Medicine Development Corporation. Dr. Lipkin is its director. Last month, the federal Department of Health and Human Services designated the center as one of eight regional biodefense centers, with $9 million a year in aid, to work on threats like smallpox, Ebola virus, plague and botulinum toxin.

This month, the Mailman School became one of about 120 institutions worldwide that make up the W.H.O.'s Global Outbreak Alert and Response Network, helping the world group investigate and contain eruptions of contagious disease. Ten other network members are based, at least partly, in the United States, including government agencies like the Centers for Disease Control and Prevention, Harvard, Johns Hopkins, and international groups like the International Federation of Red Cross and Red Crescent Societies, and the International Rescue Committee.

Mailman is ''doing cutting-edge stuff that's new for them, and it's great to have this resource in the city to call on when things hit,'' said Dr. Thomas R. Frieden, New York City's health commissioner. In an enterovirus outbreak this month on Staten Island, Dr. Frieden said, the Greene lab volunteered to help the city -- which has its own B.S.L. 3 lab -- try to identify the microbe.

Such work is not the traditional role of schools of public health. With some notable exceptions, they have tended to focus on long-term health threats, data analysis and shoe-leather epidemiology, leaving the response to outbreaks and sophisticated microbiology to government agencies, medical schools and hospitals.

But Dr. Lipkin said the lines between the two roles have been blurred, as scientists learn of links between germs and chronic diseases like schizophrenia or diabetes, as they find that formerly harmless germs can suddenly become virulent, and as a more crowded, connected world makes it easier for diseases to jump species to species and continent to continent.

''Anything that we learn about outbreaks of acute disease is going to help us in chronic diseases, as well,'' he said. ''There is no virus I know of that doesn't ultimately cause a problem. They all find their niche.''

When the SARS epidemic ended last July, experts were concerned that it would come back from wherever it hid in nature. But officials of the World Health Organization were just as worried about a new epidemic emerging from a SARS sample that escaped from one of the many laboratories working with it.

Those fears were confirmed by two accidents in Singapore and Taiwan in August and December.

Now health officials are more alarmed. A third laboratory accident in China has proved far more serious than the earlier two because it spread to three generations before the first case was detected.

The outbreak began March 25 after a graduate student became infected in China's main SARS laboratory in Beijing. Her case led to two more waves of transmission involving seven people -- including her mother, who died -- before her illness was recognized as severe acute respiratory syndrome on April 22.

Another graduate student who worked in the same laboratory fell ill on April 17, the outbreak's ninth case. He is not known to have transmitted SARS to anyone.

The students most likely became infected in separate accidents at the Institute of Virology at China's Centers for Disease Control, where some of China's top virologists work.

A World Health Organization team has joined Chinese experts in investigating the outbreak but has not pinpointed any breach, said Dr. Julie Hall, the SARS team leader for W.H.O.'s Beijing office.

''We have ruled out the obvious, and the investigation is complicated and likely to continue for a considerable period,'' Dr. Hall said.

China has sent specimens collected from patients and from various parts of the virology building to a lab in Hong Kong for independent confirmation and molecular tests that could provide clues to how the accidents occurred and how to prevent additional ones.

The two student cases bring to four the known SARS infections that have occurred in three laboratories since the disease was first detected last year. The two earlier incidents were related to inappropriate procedures in laboratories in Singapore and Taiwan, and did not lead to additional cases.

Laboratory infections usually result from inadequate training, substandard facilities and human error. Visitors to China's C.D.C. have described the SARS lab as new but the building as dilapidated.

''It is what we really feared, that the most dangerous place is from an accident in a laboratory,'' said Dr. John MacKenzie, an Australian virologist who visited the lab in Beijing in early April, after the accident but before it was recognized.

W.H.O. recommends that work on SARS be performed in a laboratory classified as Biosafety Level 3, the second most biologically secure ranking.

Dr. MacKenzie said that although the specifics of the China incident were not known, it stressed ''the need for a common definition of what a B.L. 3 lab is and also a common set of criteria under which B.L. 3 labs are inspected and are shown to be up to scratch.''

''One problem is what happens when you have B.L. 3 and B.L. 4 laboratories in old buildings in developing countries,'' he went on, adding: ''What struck me as quite bizarre is, here is a student working in a laboratory that handles SARS, and when she gets a respiratory disease and pneumonia, it takes three weeks before they even tested for SARS. That is a long period of time.'' In that period, for example, she could have infected fellow passengers on trains.

Dr. Hall said that if doctors had asked the student about her work and whether other members of her family were sick, ''alarm bells might have sounded earlier.''

Another troubling fact is that students have been involved in three of the four SARS laboratory incidents.

The Chinese outbreak, Dr. MacKenzie said, ''just adds fuel to the need for common international standards for what kind of training we have for laboratory workers.''

''Obviously,'' he continued, ''if you have good laboratories and well-trained people, the chance of cases from laboratory accidents diminishes enormously.''

Accidental infections at United States government laboratories have been rare in recent years.

In the last five years at the Centers for Disease Control and Prevention in Atlanta, two staff workers developed Lyme disease while working with ticks, and a third developed West Nile fever from a needle stick.

At the National Institute of Allergy and Infectious Diseases, the director, Dr. Anthony S. Fauci, said that from 1982 through 2003, accidents were few: a worker developed a bacterial pneumonia, and four others were infected with hepatitis A and tuberculosis without becoming ill.

The secret, he said, is formal courses to give new workers intensive training before they can handle dangerous infectious agents. ''The virologists who get nipped,'' he went on, ''are those who run experienced labs and then have a brand new postdoc or worker who comes in and gets into difficulty.''

Influenza virus infection has probably shaped human populations for centuries, if not millennia. Novel influenza viruses formed by genetic reassortment of avian and mammalian viruses emerge sporadically and, if they have the necessary infectivity and transmissibility in humans, spread rapidly around the globe causing a pandemic. While mortality and morbidity varied widely between the pandemics of the last century, the loss of an estimated 50 million lives in the most devastating pandemic of 1918–1919 has had a lasting global impact. Here we briefly review the history and effects of influenza pandemics on the global human population and events of the time. Then we discuss some of the ways in which the experience of the 1918–1919 and later pandemics has influenced development of international influenza surveillance and global public health policy, the full impact of which will become apparent in future pandemics.

Early history of influenza

Influenza viruses evolved in aquatic birds. They remain there in a number of antigenically diverse forms as an enzootic reservoir1 from which they occasionally cross into and become established in other species, including humans. Swine and horses are known to have been infected with avian influenza viruses and suggested as intermediate hosts for introduction to humans1,2. Their domestication together with that of ducks, thousands of years ago, could have provided opportunities for repeated introductions into humans.

Outbreaks of human influenza have been chronicled from the late Middle Ages in Europe and Britain, and although outbreaks as early as the 5th century in Greece have been described by some historians  most believe that descriptions predating 1520, and possibly later, should be treated with caution3,4. Many of these outbreaks were accompanied by high morbidity and sometimes mortality. However, it is unclear whether any of those recorded in the pre-Elizabethan period shaped world history although influenza reputedly contributed to the demise of Oliver Cromwell2. A possible exception to this was the suggestion that influenza was introduced to the Americas by Columbus’s second voyage in 1492, and was the first of the Old World diseases to depopulate indigenous peoples in Hispaniola and beyond with many hundred thousand deaths5. Although this has been questioned, it seems probable that influenza contributed to the depopulation of the New World and other remote, immunologically naive communities including French Polynesia6. Similarly, the first recorded influenza outbreak in Australia in 1820 severely affected the indigenous population7. A disproportionate impact on indigenous populations in Australia8, New Zealand9 and elsewhere10 persists to this day.

Pandemic influenza: the 1918–1919 pandemic

We now know that recent, and presumably previous, influenza pandemics involved viruses bearing surface antigens, haemagglutinin (H) and neuraminidase (N), to which most or all of the population lacked immunity. These surface antigen genes originate from the avian influenza gene pool, by adaptation or genetic reassortment with circulating human or animal influenzas, thus acquiring the ability to infect and be serially transmitted in humans. Spread is global and rapid, with high morbidity and often high mortality. The new pandemic virus then persists for decades as a seasonal infection by virtue of exceptional mutability.

Although numerous epidemics with significant morbidity and mortality occurred from the 16th century it is generally agreed that the first recorded outbreak meeting pandemic criteria is that of 1580 followed by those commencing in 1729, 1781, 1830 and 18893,4. The pandemic of 1918–1919 was the most deadly influenza pandemic recorded with a recent estimate of 50 million or more deaths11. The pandemic was inextricably intertwined with World War I. It influenced the capacity to conduct hostilities12 while its evolution may have been influenced by crowded conditions and soldiers’ exposure to toxic gases13. Remarkably the genetic blueprint of the H1N1 pandemic virus has been determined and the virus reconstructed14. Nevertheless, unanswered questions remain: how and where the virus evolved and reasons for the unusual age distribution of deaths and usually three distinct waves of differing pathogenicity.

Mortality varied across countries with sparing attributed to affluence15 and delayed introduction achieved by maritime quarantine16. Australia, where introduction was delayed until March 1919, had one of the lowest mortality rates, particularly in Tasmania16. Undoubtedly the pandemic had enormous social and economic impact in most societies but most attention was given to elucidating its cause, mode of transmission and toll, rather than to its social and cultural dimensions or consequences. It may never be possible to disentangle these from the consequences of the war. In Australia, where it stressed relations between the states, it was a catalyst for formation of the Commonwealth Department of Health17. There is, however, the intriguing possibility of one huge historical consequence. US president Woodrow Wilson suffered severe influenza absenting him from much of the WW1 armistice negotiations. This resulted in Britain and France imposing severe punitive conditions on Germany in the Treaty of Versailles which may have facilitated the rise to power of Adolf Hitler18.

The virologic era

Influenza viruses were first isolated in the early 1930s, and shortly afterwards the first inactivated virus vaccines were produced and used by the allied forces in the Second World War. In 1947, a major antigenic change in the circulating Type Aviruses resulted in vaccine failures. This kindled fears of another influenza pandemic like 1918–1919 and was a driver for an informal meeting of influenza experts during the 4th International Congress of Microbiology in Copenhagen in July 1947. They recommended to the Interim Commission of the World Health Organization (WHO) that an international surveillance program for influenza be initiated. Following adoption by WHO in September 1947, a World Influenza Centre was established at the National Institute for Medical Research, London, to work with regional laboratories and, later, national centres to study epidemiology and isolate new influenza strains19. In Australia, the Commonwealth Serum Laboratories were designated as a Regional Influenza Centre in 1951, upgraded to Collaborating Centre in 1992, until 2006 when the responsibility for the Centre was transferred to the Victorian Infectious Diseases Reference Laboratory. Today, there are 141 national centres in 101 countries and six WHO Collaborating Centres for Influenza in London, Atlanta, Melbourne, Beijing, Tokyo and Memphis, the latter concerned with the Ecology of Influenza in Animals. The efficacy of the network in surveillance was demonstrated during the 1957 and 1968 pandemics and the re-emergence of H1N1 in 197720.

The pandemics of 1957 (Asian) and 1968-69 (Hong Kong) were far milder than that of 1918–1919. The first was moderately severe globally and spread quickly, reaching Australia within 3 months of its origin in China21, while the second was a ‘smouldering’ pandemic and had a delayed peak in many countries including Australia in 197022. In addition to demonstrating the difficulty in producing vaccine in the necessary timeframe23, these pandemics provided the viruses which allowed elucidation of the origins of pandemic viruses.

The Fort Dix episode

In 1976 a small influenza outbreak occurred among military recruits at Fort Dix, NewJersey, with severe respiratory disease in 13 soldiersand one death. The outbreak involved a combination of H3N2 virus and a virus related to early swine-like H1N1 strains derived from the 1918–1919 pandemic24. Concerned that it might signal a re-introduction of a 1918 pandemic-like virus, and in a divisive decision, the US Government planned to immunise the whole country; scientific and medical evidence did not support the decision. The program was stopped after an epidemiologic association between vaccination and increased incidence of Guillain-Barré Syndrome was reported, something which has not been significant in subsequent surveillance25. The incident prompted the beginnings, in 1978, of formal pandemic preparedness planning by the USA and a handful of other countries, although WHO did not release its first pandemic planning document until 1999.

Avian influenza and recent pandemic concerns

The 1950s discovery that fowl plague was due to an influenza A virus, followed by the isolation of many additional avian influenza viruses, revealed the role of birds as a virus reservoir and the multiplicity of virus subtypes. H7 (responsible for fowl plague) and H5 can be highly pathogenic in domestic poultry, with exceptionally high mortality, and outbreaks of both subtypes have been recorded around the world, including Australia, since the 1950s26. Before 1997, recorded transmission of avian influenza to humans was rare, restricted to H7 viruses and almost always mild27.

In 1997, 18 H5N1 cases in humans with 6 deaths signalled poultry infections in Hong Kong. Culling appeared to stamp out the virus; however it reappeared in Thailand and Vietnam in 2003, and has since spread widely in birds and caused over 660 human cases with morethan 390 deaths across 16 countries28. Global concern over the pandemic potential of H5N1, so soon after SARS, reinvigorated pandemic planning nationally and internationally. An International Partnership on Avian and Pandemic Influenza was forged at a highlevel Plenary Meeting of the United Nations (UN) General Assembly in 2005, and the UN Secretary-General, Kofi Annan, subsequently pledged that theUNwould do all it could to ensure all countries, rich and poor, were protected and prepared for an avian influenza pandemic. A UN Systems Coordinator for Avian and Human Influenza (UNSIC) was appointed and International Ministerial Conferences on Avian and Pandemic Influenza (IMCAPI) were convened. These provided the impetus for a global response to emerging diseases and a One Health approach to manage zoonotic diseases29. In Australia, the National Action Plan forHumanInfluenza Pandemic and the Australian Health Management Plan for Pandemic Influenza have continued to be refined30,31.

Strain sharing: a new challenge for international collaboration

The H5N1 zoonotic also prompted the Government of Indonesia in 2006 to raise an issue of international equity. Developing countries share potential pandemic influenza viruses, such as A(H5N1), with WHO to support the development of vaccines and other interventions but cannot afford to purchase those products for the protection of their own people. Indonesia stopped sending H5N1 viruses to international WHO laboratories in protest. Several other countries lent their support to the campaign.

In response, WHO convened a series of technical, intergovernmental and working group meetings between 2007 and 2011. Discussions were technically, legally and politically complex, spanning influenza virology, commercial vaccine development and production, intellectual property, and sovereign rights of nations over their biological materials32. Ultimately, the meetings produced a new framework, the Pandemic Influenza Preparedness Framework for sharing of influenza viruses and access to vaccines and other benefits, adopted by the World Health Assembly in 201133,34, a crucial feature of which is the sharing of vaccines, antiviral drugs and other benefits through donations and financial contributions to WHO by manufacturers.

Implementation of the Framework faces challenges, including the use of genetic approaches to vaccine development that bypass the need to share the viruses themselves. Regardless, the Framework has embedded the principle of equity in the activities of theWHO’s Global Influenza Surveillance and Response System. The true test will come with the next pandemic.

First pandemic of the 21st century

In March 2009 a respiratory outbreak in Mexico was caused by a reassortant swine H1N1 virus more closely related to pre-1950s viruses than to recent strains. With initial indications of high mortality in Mexico and rapid spread through North America and beyond, the WHO issued escalating alerts until 11 June when a pandemic was declared. Many countries, including Australia, had already activated their pandemic plans and placed orders for many millions of doses of vaccine. In retrospect mortality was well below the early indications from Mexico and, while virus spread and major target groups were generally similar to previous pandemics, overall impact was moderate. Vaccine availability again lagged behind the peak of the outbreak, invigorating efforts to develop “universal” influenza vaccines that will protect against diverse subtypes, as well as novel delivery systems and faster production processes35. There were allegations that the WHO and national responses had been unduly influenced by commercial interests and it became clear that planning documents lacked the flexibility to tailor responses according to ongoing assessment of severity36. While these issues have been largely addressed, pandemic preparedness and public support may have been damaged37.

The next pandemic

Today, H5N1 remains widespread in poultry and human infections continue. This large family of viruses has diversified genetically and antigenically, to the extent that a single vaccine would not protect against all circulating strains. In some H5N1 strains, only a few further mutations may enable respiratory droplet transmission between mammals38.

In 2013, a new threat emerged with human infections (>440) and deaths (>120) due to a novel H7N9 avian influenza in China39. Although this virus, unlike H5N1, causes a silent infection in poultry, human infections are strongly associated with exposure to live bird markets. These viruses have several features of adaptation to mammalian infection.

Although these and other novel influenza viruses pose a continuing pandemic risk, China’s response to H7N9 exemplifies the benefits of the international networks and frameworks developed as a result of the 1918–1919 pandemic and later events outlined here. Early announcement of the outbreak, sharing of sequences and viruses, and openness about the epidemiology enabled rapid preparation of diagnostic reagents and candidate vaccine viruses by WHO laboratories and early interventions to limit spread. While sporadic influenza pandemics are an inevitable companion tohumanhistory, such international cooperation is the key to minimising their future impact on its course.

Acknowledgements

The Melbourne WHO Collaborating Centre for Reference and Research on Influenza is supported by the Australian Government Department of Health.