Photo Essay: Medical Science's Battle Against Influenza
The influenza virus The most virulent form of the flu virus (seen here under an electron microscope in a tissue sample), known as type A influenza, consists of nothing more than 8 strands of RNA, enclosed in protein capsules, surrounded by a globular envelope of proteins.
Like all viruses, influenza needs to take over a host cell in order to reproduce. Two proteins -- hemagglutinin, or HA, and neuraminidase, or NA -- found on the surface of the virus bind to proteins on human and animal cells and act as the chemical "keys" letting influenza enter and take over those cells.
Researchers have identified 16 major types of HA protein and 9 types of NA protein circulating among flu viruses in nature, and refer to influenza subtypes by which types of proteins they include: thus a flu virus whose envelope includes type 2 HA protein and type 3 NA protein is known as "H2N3."
Two other types of influenza, B and C, exist, but do not exhibit the same sort of variation.
Credit: Centers for Disease Control/C. Goldsmith, J. Katz, and S. Zaki
Evolution on fast forward RNA viruses like influenza are prone to mutation (DNA-based lifeforms are much better proofreaders), and type A influenza has become a master shapeshifter, depending on its instability for its very survival. Since animals that survive viral infections develop anitbodies that prevent further attack by the same virus, it's only by changing the nature of its surface proteins that influenza can keep tricking its hosts' cells into letting it in year after year. The virus itself can change so quickly that it becomes unrecognizable to the immune system of animals over the course of a single season. That's why humans -- and influenza-prone animals, like this chicken -- need a new flu vaccine each fall.
Influenza can mutate in response to environmental changes (exposure to antiviral drugs, for example) but it also changes its genetic identity as a matter of habit as it replicates within cells. If two different strains of influenza virus attack the same cell, they end up "sharing" in the takeover of the cell's reproductive machinery, and in the process of producing new virus particles they end up sharing genetic information. If they simply exchange a few amino acids, the process is known as recombination. If the two strains swap entire RNA strands, the process is known as reassortment.
Small changes in the flu's genetic code are known as antigenic drift, and it is the constant drift in flu viruses that accounts for yearly seasonal epidemics. Wholesale reshufflings of the virus (such as the swapping of one HA- or NA-protein for another) are referred to as antigenic shift. While antigenic drift within a flu subtype can result in wide variations in virulence, antigenic shifts are thought to be the events that trigger pandemics, since they can result in entirely new subtypes that can "jump" to new species which have no natural immunity to their combinations of surface proteins.
Birds, pigs, and people While some subtypes of influenza A -- those including the H1, H2, or H3 surface proteins -- are common in human beings, other types, including H5, H7, and H9 are thought to be primarily diseases of aquatic birds such as ducks, geese, and swans. Other subtypes affect a wide range of animals, including horses, pigs, and whales. Following antigenic shifts, animal subtypes of the flu can occasionally cross over to humans. Such illnesses are known as "zoonotic" diseases.
Wild waterfowl -- ducks in particular -- serve as the flu's natural "reservoir," the place in nature where the virus exists between outbreaks in other species. There appears to be significant species-hopping among the flu subtypes that infect birds, and animal domestication, by bringing species together in close proximity, has provided a perfect environment for such species crossovers to occur -- and to take hold among a closely-packed captive animal population.
In most cases, avian flus -- among them some of the most destructive flu viruses -- are considered unable to infect human beings directly. Generally, it has believed that a third, intermediate animal host must serve as a mixing chamber. Pigs, which can become infected with both avian and human flu viruses, have been considered the most likely candidate; the thinking being that if a pig suffers a double infection with two flu viruses, the viruses can reassort, with the emergent new virus exhibiting both avian virulency and an ability to infect humans.
Credit: Micah Fink
H5N1:The makings of a pandemic The H5N1 influenza virus, which was first identified in South African terns in 1961, is classified as a "highly pathologic avian influenza," or HPAI -- a variant of the flu defined by its ability to cause severe outbreaks of lethal disease in bird flocks.
The presence of the H5 or H7 surface protein generally indicates an HPAI, and H5N1 is no exception. Since the current strain emerged in Hong Kong in 1997, it has mutated a number of times, with a variant known as Z+ proving to be incredibly virulent, killing almost 100 percent of infected chickens. It has also shown a disturbing proficiency at infecting a wide range of mammals, including cats, which are generally considered resistant to influenza.
H5N1 also has the unusual ability to attack humans directly, often with fatal results. Since 1997, it is suspected to have infected hundreds of people, and killed at least 62 as of early September 2005. In humans, it can be incredibly destructive. While most common flu deaths are caused by opportunistic infections (like bacterial pneumonia, which takes advantage of an immune system weakened by the flu), the current strain of H5N1-- like the 1918 pandemic strain -- is powerful enough to kill directly, either through infection of the nervous system or by viral pneumonia (as seen in the X-rays above -- the white clouding indicates progressive cellular damage to the lungs of a flu patient), which results in an overreaction from the immune system that destroys the lungs.
Given this virus subtype's potency, it is feared that if a recombination or reassortment gives it the ability to move easily between human beings, it could trigger a massive pandemic -- and for some scientists, the question is not if a pandemic will erupt, but when. If the virus maintains its virulency, the death toll could rival or exceed that of the 1918 "Spanish" flu, potentially causing tens of millions of deaths.
Credit: Centers for Disease Control
On the front lines Vietnam, with the largest number of reported cases (at least 90) and deaths (40) as of September 2005, is the center of human H5N1 infection. Pictured here at far right is nurse Nguyen Duc Tinh, who is suspected to have acquired an H5N1 infection directly from a patient, Nguyen Sy Tuan (seated, wearing stethoscope). Both men, along with Sy Tuan's sister, Nguyen Thi Ngoan, survived the illness (Their stories are told in the film H5N1 · Killer Flu)
Most of the human H5N1 infections have been attributed to direct contact with sick poultry, but this group of cases (along with another so-called "family cluster" of cases reported in Thailand), in which H5N1 may have been transmitted from person to person, makes a pandemic seem a real possibility. The virus was not lethal in this particular cluster of cases, but some epidemiologists feel that the lack of lethality may not be a cause for celebration, but instead an indication that the virus is adapting to human hosts and perhaps a portent of much larger numbers of cases to come -- with many more deaths to follow.
The death rate from human H5N1 infection has been falling over the past year, though that may be because mild cases are now being recognized when they had not been in the first years of the outbreak. The meaning of these cases is still far from clear.
Credit: Thirteen/WNET/Blue Ice Pictures, Inc.
Tracking the disease Here, an Indonesian research technician examines chicken tissue samples as she prepares to test for the presence of H5N1. Keeping close watch on the disease's continued spread in the domestic and wild bird populations is essential at this point, as is acting quickly to contain outbreaks among chickens in order to avert severe economic impacts and to minimize the possibility of spreading the disease to poultry workers. Across Asia, some 200 million chickens and ducks have been killed outright by the disease or culled in an attempt to stave off further deaths, resulting in massive losses for poultry producers large and small.
More importantly, in the absence of a proven treatment or vaccine, it is fundamentally important for the future of human health to recognize H5N1 outbreaks as soon as they appear, if only to buy time for the development of effective medical and public health measures.
Since H5N1 influenza is a zoonotic disease, communication between animal and human health specialists is key in tracking the virus and planning a response to it. In previous pandemics, humanity had no access to detailed advance information about its viral enemies. H5N1's attack on the avian population may be horrifying, but it may well have lessons for human researchers looking to formulate a medical response. If policymakers can be convinced to join the effort, it is at least possible to begin combatting an H5N1 pandemic before it has a chance to begin.
Credit: FAO/A. Ariadi
The public health response Official responses to the rise of H5N1 have varied widely: the city of Hong Kong (under the direction of Margaret Chan, who now heads the World Health Organization's avian influenza program) moved quickly to cull its bird population; on the other hand, it appears that the Chinese government (repeating its behavior during the 2003 SARS outbreak) was slow to acknowledge it had an H5N1 problem at all, and may have failed to report human and animal cases and deaths. The responses of other nations have fallen along a spectrum betwen the two.
H5N1 recognizes no national boundaries (see the Interactive Map for a closer look into the spread of the disease), so information exchange between nations may be one of the only effective tools in slowing viral exchange. Unfortunately the WHO reports that only 40 of its 192 member nations have prepared guidelines establishing pandemic preparedness.
Those countries that do have plans in place still face the difficult project of preparing for a pandemic that, should it happen, would likely overwhelm the hospitals of even the wealthiest nations. Governments are attempting to stockpile what vaccines and antivirals do exist, ongoing research is achieving some useful results, and H5N1 pandemic preparedness was a subject of discussion at the 2005 United Nations World Summit.
Still, more can be done immediately on the animal management front. Pictured are Vietnamese public health workers disinfecting a chicken house following the culling of sick birds, a common response to the discovery of H5N1 outbreaks, along with the monitoring of bird transport and the disinfecting of vehicles carrying birds. Many nations have also begun vaccinating their chicken populations (a number of tested animal-only vaccines agains H5 influenzas, including H5N1, now exist).
Credit: Thirteen/WNET/Blue Ice Pictures, Inc.
The path to a vaccine Since the flu circulates seasonally among birds and humans, an industry is in place that prepares a vaccine each year. Ordinarily, the vaccine is prepared by assessing the three flu subtypes likely to be a threat in a given year. A number of monitoring stations, overseen by the World Health Organization, make the determination. Researchers grow samples of those viruses in chicken eggs, inactivate the viruses with chemicals, and then harvest their surface proteins, which are mixed into a vaccine -- the yearly flu shot. The vaccine produces immunity by fooling the human body into producing antibodies to those combinations of proteins, and thus to the virus subtypes that include them.
The problem with vaccine approach to preventing flu pandemics is that the influenza virus is a fast-moving target. By the time a vaccine is in production (the production process can take up to nine months), the prevailing flu viruses in circulation may have changed so much that any antibodies developed in response to the vaccine may be useless. The process is also labor-intensive and limited by the number of available eggs. Even in a high-production year, only some 300 million doses of vaccine are made -- not nearly enough in the case of a serious pandemic which might infect billions.
H5N1 presented vaccine researchers with an even bigger obstacle. Since its most virulent forms are 100 percent lethal in chickens, traditional methods of preparing a vaccine proved unusable. After some experimentation, a vaccine based on a modified strain of H5N1 that can be grown in chicken eggs has been developed and is currently in clinical trials; the early results are promising, but its effective dosage thus far appears to be extremely high, so it is unclear whether enough vaccine can be produced to protect a significant number of people.
Credit: Micah Fink
The frontiers of medicine Recent technological advances may be able to speed up the vaccine production process. One approach abandons the use of chicken eggs in favor of growing the vaccine-precursor viruses in an artificially produced "cell culture."
Another process, in use by a team led by flu specialist Robert Webster of St. Jude Children's Research Hospital in Memphis, TN, involves using genetic engineering techniques in an attempt to "build" specific flu vaccine proteins amino acid by amino acid, allowing quick custom tailoring of vaccines to resemble circulating strains. Currently, since vaccine production harvests proteins from influenza viruses captured in the wild, by the time the vaccine is made available, wild viruses may have mutated so much that the vaccine is no longer useful.
Experiments are continuing into a "universal" flu vaccine that would ignore the variable HA and NA surface proteins and instead induce immunity against some of the more stable proteins in the influenza virus. The British firm Acambis has been conducting animal trials of such a vaccine, but use in humans may still be years away.
Antiviral drugs have shown promise in treating diseases such as HIV/AIDS, and there are several antiviral agents that are at least theoretically effective against type A influenza strains. Unfortunately, the Chinese government seems to have encouraged the use of one of the antiviral agents, amantadine, on threatened chicken flocks over the past few years, and as a result the current variant of H5N1 is resistant to the drug. Another antiviral, oseltamivir (manufactured by Roche under the name Tamiflu) is effective against H5N1 in the lab, but its usefulness against pandemic influenza is unknown.
Research into these methods continues, often in novel ways. Above we see a computer-generated model of a space-grown, highly purified crystal of the influenza enzyme neuraminidase, used in the design of antiviral drugs.
Credit: NASA Marshall Space Flight Center