Wednesday, August 02, 2006

Study shows hybrid of human-bird flu viruses didn't transmit well in animals

Study shows hybrid of human-bird flu viruses didn't transmit well in animals
19:07:33 EDT Jul 31, 2006

(CP) - U.S. scientists who artificially engineered viruses containing genes from both the worrisome H5N1 avian flu virus and one that causes human flu found the resulting offspring failed to achieve the one feature H5N1 lacks to trigger a pandemic - transmissibility.

The study, published Monday, reports that the resulting viruses did not transmit between ferrets, animals that are susceptible to human and avian flu viruses and which are commonly used as a laboratory surrogate for human infection.

The work, done at the U.S. Centers for Disease Control in Atlanta, may provide clues as to why H5N1 has not ignited a pandemic despite having circulated in Asia since at least 1996. To date the virus has infected at least 232 people in 10 countries and has killed at least 134 of those people.

But the study findings cannot be used to predict whether H5N1 is capable of acquiring the characteristics to easily transmit to and among people, the head of the CDC cautioned.

"These data do not mean that H5N1 cannot develop into a pandemic strain," Dr. Julie Gerberding told reporters in a telephone briefing before the paper was released by the journal the Proceedings of the National Academy of Sciences.

"Flu is always unpredictable. And though we weren't able to do this through some simple gene exchanges . . . there are many other combinations and subtle changes that the virus itself could make," Gerberding said, noting she is concerned the findings may lead people to conclude the worry over H5N1 has been overblown.

A number of leading flu scientists who were not involved in the research concurred with Gerberding's take.

"I think we would be foolish to assume that H5N1 cannot do this. It may not. But that does not mean we can rule it out," said Malik Peiris, an influenza authority at the University of Hong Kong.

The CDC scientists, led by senior author Jacqueline Katz, conducted what are called reassortment studies.

They artificially engineered flu viruses that contained different combinations of genes from an H5N1 virus recovered in 1997 and H3N2, one of two flu subtypes that currently circulate during the human flu season. The scientists did not work on the other circulating human flu subtype, H1N1.

There are 56 potential combinations of two viruses, though the CDC scientists only tried to engineer about a half dozen different hybrids, those felt to be the most likely to produce troublesome combinations. Katz said to try all 56 would have amounted to "years of work."

Some gene combinations failed to grow at all. And when the scientists infected ferrets with the combinations they succeeded in creating, the resulting infections rang no alarm bells.

Viruses with a human hemagglutinin and neuramindase (the H and N in a flu virus's name) and six genes from H5N1 infected ferrets but didn't pass well from an infected ferret to uninfected animals in nearby cages.

Viruses with H5N1's hemagglutinin and neuraminidase built onto human flu genes infected ferrets but did not pass at all to nearby, uninfected animals.

Flu viruses mutate constantly and it's not known whether more contemporary H5N1 viruses would come together with H3N2 viruses more readily, or with more alarming results.

"I want to stress that our results can't be generalized and they're only relevant for the viruses that we used in the study," Katz, a branch chief in CDC's influenza division.

Nor do they indicate how likely or unlikely H5N1 is to swap genes with H3N2 or H1N1 in nature, or how transmissible such naturally occurring reassortant viruses would be.

"I think it's an important paper (but) it doesn't tell us anything about the propensity to reassort, because the reassortants were made artificially," Peiris said.

"It's really a question of: If you do have a reassortment, will it transmit efficiently human-to-human, purely by the fact of the reassortment? And what they've found - to the extent that they've done it - is No."

To test what might happen in nature, scientists would have to conduct what are called classical reassortment studies, where H5N1 and human flu viruses are allowed to co-mingle in a lab dish, producing naturally occurring offspring viruses. CDC is currently conducting such work with more current H5N1 viruses, Katz said.

The results could be different, said Dr. Frederick Hayden of the World Health Organization's global influenza program.

"We may see surprises and different gene reassortant patterns there. But until you actually do the experiments, that would just be a matter of speculation," said Hayden, who echoed Gerberding in saying this study doesn't offer any insight about how likely or unlikely H5N1 is to cause a pandemic.

There are two ways a pandemic virus can emerge, by reassortment with another virus or by slowly mutating to acquire the ability to spread from human to human. This work offers no clue as to whether H5N1 could develop into a pandemic strain through the latter mechanism.

Interestingly, the attempted combinations included viruses modelled on the two that caused the last two pandemics, in 1957 and 1968 - both of which were triggered by viruses containing human and avian flu genes. But multiple tries failed to produce a viable living virus.

"Combinations that worked to create pandemics in '57 and '68 weren't the obvious combinations that we could achieve in this current study," Katz noted, adding this suggests that for reassortment, there may not be general patterns that hold true across subtypes.

"Probably it's a constellation of effects" that would result in reassortment, she said. "And you just have to hit upon the right combination."

To test the transmissibility of the viruses, the CDC scientists devised an elaborate system of exposing infected ferrets to non-infected ferrets in nearby cages. They also tested the system to determine whether a number of flu viruses, avian and human, provoked the same type of illness - moderate, severe, lethal - in the ferrets as they did in humans.

The results suggest this model would be an effective way to quickly test the transmissibility of a new reassortant virus, if it arose in nature, the authors reported.

Adolfo Garcia-Sastre, an influenza microbiologist at Mt. Sinai School of Medicine in New York, said the model would be a useful tool to gauge the pandemic threat of future reassortant viruses, and could help scientists puzzle out the molecular mysteries that dictate why some flu strains transmit well in humans while others do not.

That in turn could help in the development of drugs that could block transmission of flu from an infected person to others, he noted, "which would be a very interesting concept for a drug. It does not help the person with the disease, but it makes the disease not transmissible."

© The Canadian Press, 2006


Post a Comment

Links to this post:

Create a Link

<< Home