Harvard Public Health Review
Winter 2006





Infectious Disease

Flu Catchers

The World Health Organization has declared that an avian flu pandemic is inevitable. On the forefront of research and surveillance efforts, HSPH-trained experts are working to stave off disaster.

There are the blaring headlines about an avian influenza outbreak, and then there are the men and women behind the scenes, out of the media glare, who are actually trying to prevent and prepare for a pandemic that infectious-disease experts say is long overdue.

Scientists trained at the Harvard School of Public Health are prominent among those movers and shakers. From vantage points all along the public health spectrum, from Petri dishes to populations, they are shining light on the avian flu virus known as H5N1, the sometimes pathogenic subtype that has felled poultry from Hong Kong to Romania and, as of December 6, had infected 134 humans in five countries (69 of whom have died). Some are tracking the biological evolution of the virus in waterfowl to see how it affects its hosts--as it prepares, perhaps, to make an easy leap to humans. Others are building sophisticated surveillance systems to halt H5N1 in its tracks. And still more are exploring the way the virus replicates in humans, the first step to developing cures. They are front and center, left flank and right, in the battle against a foe that the European Union has pledged $35.7 million to defeat in Asian nations, and that President George W. Bush has pledged $7.1 billion to defeat at home.

Pandemic influenza is a very particular beast. A flu that qualifies for the pandemic label must meet three criteria: The virus that causes it must be a new subtype; it must infect humans; and it must be able to spread easily and sustainably among humans. "Pandemic influenza is the global spread of a new strain of influenza for which few people, if any, have pre-existing immunity," explains Marc Lipsitch, associate professor of epidemiology. Lipsitch co-authored an analysis in the December 16, 2004, issue of Nature on the transmissibility of the 1918 flu virus, a bird flu that jumped to humans and killed some 50 million people worldwide. "H5N1 is widespread. It can infect humans, although it cannot transmit easily among humans. We now see what may be highly favorable conditions for the appearance of a new pandemic."

Here is how three HSPH alumni and one former postdoctoral fellow--in labs and government agencies around the world--are taking action against the threat.


It was the case that alarmed the world.

On September 11, 2004, an 11-year-old girl in Thailand died from avian flu after contracting the disease from poultry at her aunt's home. Her mother, who lived in a distant city, had come to the hospital to care for her; she, too, died of pneumonia 12 days later. The aunt, who had also ministered to the girl in the hospital, became ill but recovered.

This series of events provides the most direct evidence to date of probable human-to-human transmission of pathogenic H5N1. Prasert Auewarakul, MD--a postdoctoral fellow from 1995 to 1997 in the lab of Tun-Hou Lee, HSPH professor of virology in the Department of Immunology and Infectious Diseases--played a key role in bringing the evidence to light.

From his post as a member of the medical faculty at Bangkok's Siriraj Hospital, Auewarakul, with colleagues, analyzed viral specimens from the three victims, comparing the genetic makeup of the isolates and how the virus had interacted with antibodies from each person's blood. The researchers also meticulously reviewed the epidemiological findings of Thailand's bird-flu surveillance team, noting when each subject became sick and constructing a timeline of how her illness progressed, her level of exposure to poultry, and the nature and timing of contact between the parties. Their landmark conclusion was published in the Jan. 27, 2005, issue of the New England Journal of Medicine: H5N1 had most likely passed directly from the infected girl to her relatives.

"Once you identify a cluster like that, you have to be very vigilant about containing the spread," said Auewarakul, speaking by phone from Bangkok, where he's vice-chairman of the Department of Microbiology at Mahidol University. "Because theoretically, if you let the virus transmit from one person to another, and then to another, it will eventually adapt to humans. The danger then becomes that it will adapt to the point where it can be transmitted efficiently among humans."

Up until the outbreak of pathogenic H5N1 in Thailand, in January 2004, Auewarakul had been immersed in the work that he began at HSPH: investigating the evolution of the external characteristics of various subtypes of the AIDS virus, HIV, as their genes interacted with the environment. But with few laboratories in the country able to respond to the bird-flu threat, he had to quickly change gears.

Today his research extends in two directions: In collaboration with two veterinary schools, he's monitoring any changes in H5N1 in order to track its virulence in animals. And he's concentrating on how the virus replicates in humans--what characteristics enable it to infect human cells, which organs it replicates in (lungs and intestines have been identified so far), and how it causes disease. One area that particularly intrigues him is the role that immune-system cells called cytokines play in H5N1's pathogenicity.

The body fights off infection not only with antibodies (that is, if it recognizes the intruder) but also with cytokines--small proteins released by immune and other cells that affect how cells behave and interact. In the best-case scenario, cytokines can stop a virus from replicating. But their ferocity in battle can have paradoxical effects as well, as one side effect of cytokine activity is inflammation.

"There's a hypothesis that this virus is able to induce human and animal cells to produce an extraordinary number of cytokines while still being able to replicate," explains Auewarakul. It ignites a cytokine blitz, which in turn causes extensive inflammation and tissue damage. This "hyper-induction" of cytokines may be one reason, he says, that the infection is so virulent: The body is essentially wreaking havoc on itself, contributing to its organs' vulnerability. "Understanding how the virus induces cytokines--and survives against the cytokine defense--could provide us with ways both to minimize tissue damage and to control the virus's spread," says Auewarakul.

Plans are under way to establish a collaboration between Auewarakul's lab, at Mahidol University, and Tun-Hou Lee's, at HSPH, to monitor changes in the genetic makeup of H5N1 and to analyze how such changes may affect flu-vaccine design and development. "With the knowledge we're gaining, hopefully we'll be able to prevent a chain of human-to-human transmission, which the virus would need to adapt enough to spark a pandemic," says Auewarakul. "I'm optimistic. I don't think we're going to face a catastrophe like the flu of 1918."


In 1997, scientists in Hong Kong were the first to detect the virulent bird-flu strain of H5N1 in humans, after 18 people contracted the virus and six died. That early experience has given Gabriel Leung and his colleagues in Hong Kong (including virologist Malik Peiris, who's credited with identifying the SARS coronavirus) a head start in developing a surveillance system that might stem the spread of pandemic influenza around the world.

It was Leung's year as a Takemi Fellow in International Health at HSPH in 2005 that honed his ability to build statistical models useful for tracking not just the source of H5N1 but also the conditions that promote the virus's amplification and transmission. Leung chairs the Hong Kong Scientific Committee on Advanced Data Analysis and Disease Modeling. Every week since 1999, the committee--in collaboration with government and university virologists and veterinarians--has been collecting and analyzing data on a broad range of variables, including import and sales figures for chickens in wholesale and retail markets throughout Hong Kong, patterns of viral circulation, the precise location of infected chickens (stalls, cages, farms), and local weather conditions. The aim is to evaluate various health policy interventions, which so far have included vaccinating all imported birds before they enter Hong Kong and scheduling monthly "rest days," when all live-chicken markets are washed out and every unsold bird is killed.

"My committee is charged with helping the government collect evidence and run it through statistical models to see how Hong Kong not only can save itself but, more important for global health, be the sentinel node for the virus--that is, detect the first human cases in this part of the world, where a pandemic is most likely to start," says Leung. The spread of SARS, he points out, didn't explode until the disease hit Hong Kong, in February 2003; then, within a week, it struck 46 other countries.

"The goal is two-pronged: To warn the rest of the world that a pandemic strain exists, and to slow the strain down," Leung explains. The complex models his team is constructing will also point the way to which interventions--including quarantine, isolation, and antivirals like Tamiflu--are likely to be the most effective in so-called "mega-cities," those with 10 million or more people.

Bird flu can be contained in small rural villages, because residents there rarely travel and the number of affected birds is limited. The danger comes when the disease hits large settings, like Hong Kong, with its seven million residents.

"The theory that everybody's working on is: Detect the pandemic strain at its source and install strong defenses in mega-cities, which are likely the amplification and re-export centers," says Leung. "Then buy enough time to rev up the development of a reverse genetic vaccine, so we can quickly cover everybody.

"The job of modelers is to explore how much the virus can be slowed down, and the requirements to do that: Do we have to stop people from traveling? Do we have to close all schools and offices? Do we have to go on a state of high alert and declare martial law? How far should we go in stripping people of their liberties and rights in the name of public health?"


Wrapped in a disposable Tyvek suit, and with a battery-operated air purifier at her waist, Katharine Sturm-Ramirez, PhD '01, travels daily to the front lines in the battle against the spread of avian influenza.

Sturm-Ramirez--who trained at HSPH under Phyllis Kanki, professor of immunology and infectious diseases--is a postdoctoral research associate in the lab of Robert G. Webster, the veteran bird-flu expert at St. Jude Children's Research Hospital, in Memphis, Tenn. Webster is credited with demonstrating, over 30 years ago, that wild aquatic birds are the natural reservoirs for all influenza viruses and that pandemic strains arise when viruses in people and animals swap genes. According to Sturm-Ramirez, it was her coursework at the School in biological sciences and public health--along with her research in Kanki's lab on HIV infection in women--that gave her both the virology expertise and the epidemiological skills to land a spot on Webster's international team.

"I started working on HIV in '96–'97, at a relatively late stage of the AIDS pandemic," says Sturm-Ramirez. "I switched to avian flu largely because it offers the opportunity to prevent something big from happening--that is, if we can learn enough about this ever-changing virus."
Sturm-Ramirez's contribution to that knowledge goes to the heart--or, more accurately, the intestines, lungs, and brain--of the matter, as she tracks the biological evolution of H5N1 viruses. Her research focuses primarily on ducks, "the Trojan horse of H5N1," says Sturm-Ramirez, quoting Webster's metaphor. Historically, aquatic birds have simply been carriers of flu viruses; they spread them to chickens and other birds, which die soon thereafter, yet rarely get sick themselves. But this scenario changed dramatically in late 2002. Ducks and geese infected with H5N1 were found dead in recreational parks in Hong Kong, indicating that the virus's genetic makeup had significantly changed. Sturm-Ramirez and her colleagues analyzed the new highly pathogenic strain, and characterized it, in a paper published in May 2004 in the Journal of Virology.

"My questions became: 'Why, suddenly, did H5N1 start killing its natural host?'" says Sturm-Ramirez. "How--and how fast--will it change in the future?" And since ducks were no longer just carriers but also victims, she wondered, "What role do ducks play now in the epidemiology of H5N1 influenza?"

As H5N1 travels from ducks to chickens to wild migratory birds across Asia and Europe, Webster's lab--one of eight World Health Organization reference laboratories for bird-flu diagnosis around the world--receives samples of the latest avian killers, carefully sealed in cryogenic tubes. It's Sturm-Ramirez's job to test the new isolates in ducks, in order to determine their pathogenicity--that is, their likelihood of causing disease. Considering the risk involved in handling such lethal viruses, the work has to be performed in a high-level biological containment laboratory.

"As a scientist interested in viruses of public health importance, I can't help but be excited about working on H5N1," she says. "But working under biosafety level 3+ conditions is very intense and physically exhausting. It certainly sounds much sexier than what it actually is."
In one study, published in the September Journal of Virology, Sturm-Ramirez, the lead author, took swabs of the infected ducks' upper respiratory tracts and their intestinal tracts in order to follow the virus's trail through their bodies. She also opened up the ducks that died and explored their internal organs.

Her findings were revelatory: Previous influenza viruses have replicated mostly in aquatic birds' intestines, and have been transmitted when healthy birds drank water contaminated with sick birds' waste. But Sturm-Ramirez discovered that recent H5N1 viruses had spread throughout the birds' bodies (including their brains) and replicated at higher levels than expected in their upper respiratory tracts.

The switch could have serious implications for the transmission of H5N1 to humans. "A virus can infect a cell only when it's able to bind to the receptor of that cell," says Sturm-Ramirez. "Until now, H5N1 seemed to replicate primarily in intestinal cells in ducks. But with its enhanced ability to replicate in respiratory tract cells, it may be one step closer to efficiently infecting mammals, including humans, because the respiratory tract is where mammals harbor flu."

The pathogenicity of the H5N1 isolates, she found, ranged from harmless to highly lethal in ducks, while remaining highly lethal for other species. The harmless isolates, she concluded in the paper, would be the ones to worry about, as they "can propagate silently and efficiently" among domestic and wild ducks. They represent, she wrote, "a serious threat" to both veterinary and public health.

"Dead birds don't fly, so ducks that are infected but not sick may be playing a critical role in the current spread of H5N1," she says. "These viruses change very quickly. The more the virus spreads, the more diverse it becomes, creating a greater risk that there will be one variant that can transmit efficiently from human to human--the only step missing to cause a pandemic. It's a race against time, as we try to understand H5N1's evolution."


To date, highly pathogenic H5N1 has not made its way across U.S. borders. And part of the reason may be the rigorous poultry-surveillance infrastructure for which Saul Wilson, Jr., DVM, MPH '55, laid the groundwork in the 1980s as assistant deputy administrator for the Animal and Plant Health Inspection Service of the U.S. Department of Agriculture (APHIS-USDA).

A veterinarian by training, Wilson realized, while serving with a USDA program in Mexico to eradicate foot-and-mouth disease, that to stop an epidemic in its tracks he needed not just clinical expertise, but also epidemiological savvy. Armed with a degree from HSPH, he returned to government work, this time helping to develop cooperative state–federal programs aimed at ridding the United States of animal scourges ranging from brucellosis and tuberculosis (TB) to hog cholera and sheep scabies.

It was during this time, from 1955 to 1979, that Wilson refined his skills for analyzing disease-eradication techniques. For example, as a veterinarian in the state-federal bovine TB control program, he helped determine that annually testing a targeted number of cattle in a herd for TB and removing those found reactive wasn't stringent enough. Many infected animals were slipping through the testing cracks, identified as having TB only after they'd been slaughtered. The finding led to a shift from farm-to-farm testing to post-mortem examination. The next step entailed back-tracking TB cases to their herd of origin, in order to pinpoint which herds needed testing.

Wilson would apply this knowledge to subsequent leadership positions at APHIS. There his duties included directing staff activities for the National Poultry Improvement Plan, a joint endeavor of the commercial-poultry industry and state and federal officials to set health requirements for poultry breeding and products. At the time, the major threats to public health were salmonella and Newcastle disease. Both demanded an improved system for surveillance and monitoring.
Based on epidemiological data they gathered on the signs and symptoms of various diseases, veterinary practices in birds' countries of origin, and so on, Wilson and his colleagues designed procedures for keeping "emergency diseases" out of the country. "We developed a certification program for live birds coming here--from pets to zoo animals--as well as for chicken and turkey eggs being imported for breeding purposes," says Wilson, who received an HSPH Alumni Award of Merit in July for his contributions to regulatory policy making. Among the requirements he helped hammer out were the tests and examinations each bird had to pass to gain entry to the U.S., and the time each spent in quarantine.

"The contingency plans we set up for dealing with foot-and-mouth disease, hog cholera, avian influenza, and other exotic diseases of livestock and poultry are now being studied by FEMA for incorporation into the current bird-flu prevention effort," says Wilson. "The U.S. will need to develop vaccines for poultry and livestock, acquire hatching eggs for vaccine production, design procedures for disposing exposed but noninfected poultry, and strengthen the national network of veterinary diagnostic laboratories. So though President Bush's $7.1 billion flu-pandemic strategy is comprehensive, the funds he's requested for agriculture--$91.4 million--are out of sync with those needs."


This page is maintained by Development Communications in the Office of Resource Development.
To contact us with suggestions, comments, and questions, please e-mail: editor@hsph.harvard.edu

Copyright, 2006, President and Fellows of Harvard College