Research at The University of Hong Kong into why bird flu is such a killer in humans is focusing on chemical messengers that control the immune system, proteins called cytokines and chemokines, which are stimulated when the virus is present.
Prof Malik Peiris, leader of the research, said: “They keep the immune system under control and are good for us but if they go out of control, you get a major disaster.”
The research approach is to compare the effects of the H5N1 avian flu and ordinary flu viruses on human cells.
One major difference is that the H5N1 virus hyper-induces the activity of cytokines and chemokines which are produced by white blood cells.
Work not yet finished
“We then discovered that the chemokines were attracting macrophages which are the scavengers of the body,” he said. “Stimulating macrophages is like lighting a stick of dynamite. They are very potent cells and when they get infected they release a whole lot of other cytokines.
“What we see in the lungs of patients that die of the virus is that they are chock-full of macrophages.” The mortality rate for H5N1 patients is about 50 percent.
The next step in the research was to find out which proteins or gene segments of the virus were responsible for generating the cytokine induction. Of particular interest were the H5 and N1 genes which give the virus its name.
Levels of cytokine activation were investigated in the many different variants of the H5N1 virus that have occurred over the last 10 years and big differences were found. “The grandfather of H5N1, a virus called A / Goose / Guangdong /1 / 96 first isolated in Guangdong Province in 1996, does not actually hyper-activate cytokines,” said Prof Peiris.
The virus that caused problems in Hong Kong in 1997 had high cytokine activity. Viruses found in 2000 and 2001 had low cytokine activity but viruses from 2004 onwards are “very, very potent, even worse than the 1997 virus,” said Prof Peiris.
Using genetic manipulation techniques, the researchers made hybrid viruses between the low cytokine and high cytokine inducing viruses by re-assembling the eight pieces of genetic information that make up each virus.
When they put the H5 and N1 genes into the human virus with its known low-cytokine activity, the result was still low cytokine activity so they were able to discount H5 and N1 as suspects. In the real H5N1 virus, the two proteins appear on the surface “just like someone wearing a coat,” said Prof Peiris. “This is important for recognising the virus but the virus engine is something else.
“The work is not yet finished but what we can show is that this high cytokine property is not due to the H5 or the N1 genes but is due to the internal genetic engine of the virus.
“We know that mixing the avian and human viruses is one way that a pandemic can arise with the human virus genetic engine acquiring a new coat of H and N.
“Our work suggests that such a hybrid virus with H5 and N1 will not be a high cytokine inducing virus. You’d still have a pandemic, but the virulence may be less than in the recent human disease. On the other hand, if the H5N1 virus directly adapts to human-to-human transmission without mixing its genes with current human influenza viruses, the resulting pandemic may be very severe.”
Finding a cure may still be some way off because it seems that at least three actions are taking place. he said.
One is replication of the virus, another is the altered host response as in the hyper-induction of the cytokines and chemokines, and a third action is that the virus may also disseminate to other parts of the body.
“Treating the virus early may block the pathways of replication and dissemination but, once it has started, it will not necessarily stop the altered host response,” he said.
Prof J S M Peiris : firstname.lastname@example.org