By Dean Haycock

Special To BioWorld Today

Last year in Hong Kong, a 3-year-old boy came down with a new strain of influenza. He developed neurologic complications and was diagnosed with Reye's syndrome, a disease associated with influenza A and chicken pox infections. The child became the first victim of an avian flu outbreak that eventually killed four other persons.

Health officials worried that this new viral strain might spread around the globe, as less lethal flu viruses do every year. It is, many believe, only a matter of time before the ever-mutating and adapting viruses come up with a genetic combination that will turn what is a troublesome inconvenience for most healthy people into a lethal infection.

Most flu viruses are not deadly to healthy individuals because they can only infect cells in the respiratory tract. In part, this limitation is due to their dependence on enzymes provided by their hosts. Enzymes in the respiratory system cleave a protein called hemagglutinin (HA) on the surface of the virus. The result is two HA subunits, which allow the virus to attach to and enter the cells. The enzymes capable of processing the HA found in most strains of flu viruses are largely limited to the respiratory tract. Hence, flu infection is largely limited to cells in this part of the body.

Rarely, mutations arise in avian flu strains that affect the HA cleavage site, making it susceptible to other host enzymes. Such a mutation was detected in virus samples obtained from the young victim of the 1997 Hong Kong outbreak. While it has not been proven that this mutation allowed the virus to infect cells outside the human respiratory tract, it has been demonstrated that the mutation is responsible for systemic infection in chickens. The possibility therefore exists that the mutation might have contributed to the varied, non-respiratory signs of disease observed during the Asian outbreak.

A new model describing how changes in another viral protein might lead to more virulent strains of influenza appears in the August 18 issue of the Proceedings of the National Academy of Sciences (PNAS). Postdoctoral fellow Hideo Goto and Professor of Virology Yoshihiro Kawaoka of the University of Wisconsin School of Veterinary Medicine, in Madison, describe a novel mechanism for HA cleavage in a viral strain descended from the one that killed 20 million people in the pandemic of 1918. Nearly 60 years ago, the strain was adapted to grow in mice, in which it infects cells throughout the body. The results described in the PNAS paper, "A novel mechanism for the acquisition of virulence by a human influenza A virus," could provide a better way to predict flu hazards and provide new drug targets for countering influenza viral infections.

A crucial part of the new model is a second major viral-surface protein, called neuraminidase (NA). NA grabs and holds a host protein called plasminogen. Plasminogen is a widespread molecule that serves as a precursor for plasmin, an enzyme that can cut up proteins, including viral HA. By binding and sequestering plasminogen, NA increases the nearby concentration of a protease that can cleave HA and so allow the virus to infect cells. Because plasminogen is found throughout the body, viruses that can sequester it may be able to infect cells outside the respiratory system and do more damage.

This unusual function of NA can be traced to certain structural changes in the molecule, the presence of a lysine amino acid at one end of the molecule and the absence of an oligosaccharide side chain at a particular spot. The authors suggest that these features could provide a means "by which influenza A viruses, and perhaps other viruses as well, could become highly pathogenic in humans." Some bacteria are known to possess plasminogen-binding proteins, but this is the first example of a virus demonstrating the ability.

If the influenza strain that caused the pandemic of 1918 is ever found to have the same structural changes in its NA molecules, it might explain why it was so deadly. But Kawaoka does not feel that is the important feature of his and Goto's research.

"I don't want to emphasize too much the 1918 pandemic," he said. "I want to emphasize that our finding may prompt other investigators to look for the same mechanism in other viruses, because that is the major implication of our findings," Kawaoka told BioWorld Today.

The discovery may provide a way for public health officials to detect particularly virulent strains of flu before they have a chance to start another pandemic. "I don't think this is the only virus that uses this mechanism," Kawaoka said. Although he would not divulge those at the top of his list, the structural features that allow NA to sequester plasminogen could not be detected in several less virulent strains of human, avian and swine viruses. This finding suggests the NA mutation may be limited to virulent viruses.

"There may be other viruses causing systemic infections that may be using this mechanism already," Kawaoka said. *