Recombinomics | Elegant Evolution

Home Founder What's New In The News Contact Us

Paradigm Shift

Viral Evolution

Intervention Monitoring

Vaccine Screening

Vaccine Development

Expression Profiling

Drug Discovery

Custom Therapies


Evolving H5N1 Bird Flu Alarms Experts

Recombinomics Commentary

June 25, 2005

>> "Everything suggests, that the situation we are in now, there is a greater risk for a pandemic than for many decades," said Dr. Peter Horby, a medical officer and epidemiologist for the World Health Organization in Hanoi. "The situation is much more complex than a year ago."

"This year, there doesn't appear to be a stop," said Klaus Stöhr, head of WHO's global influenza program in Geneva. "Every human case is worrisome because there is another chance for the virus to [mutate] and a higher chance for a pandemic to occur."

Stöhr said it was unclear why human cases have not receded this summer. It could be better surveillance or more instances of bird-to-human transmission. Or perhaps the virus has become more adept at infecting people.

Michael Osterholm, director of the Center for Infectious Disease Research and Policy at the University of Minnesota, sees the developments in Asia as more reason to worry. He doubts the past will predict the virus' future.

Unlike previous pandemics, where a virus underwent a major genetic overhaul all at once, Osterholm said the avian strain has been changing gradually since it was first identified in 1997. He believes the virus will continue to transform, increasing the likelihood it will ultimately lead to a pandemic.

"We haven't done much to eliminate the source in Asia," said Osterholm, a former bioterrorism special adviser to the current Bush administration. "And there is a dynamic mutation laboratory over there. I see nothing to slow down the mutations." <<

The above comments reinforce the notion that H5N1 is changing and becoming more genetically complex, but the comments fail to reflect the fact that these changes are being driven by recombination and not new mutations.

WHO has been relying on two basic tenents of influenza genetics, but both are incorrect.  One maintains that dramatic changes (shifts) are caused by reassortment and the other maintains that small changes are caused by new mutations (drifts).  Although both mutation and reassortment play a role in influenza evolution, recombination is the true driver of rapid evolution, which is what H5N1 is currently doing,.

The above comments begin to acknowledge that H5N1 has evolved since 1997 without acquiring a human gene via reassortment.  The lack of the acquisition of a human gene was announced in virtually every press release WHO issued last year after a new isolate was sequenced.  However, the news was not news because the concept that a pandemic strain will evolve from H5N1 via reassortment with a human influenza doesn't make sense.  H5N1 in 2004 lacked the ability to efficiently transmit between humans and the human receptor binding domain is on HA.  Thus, acquisition of that domain via reassortment would require that the H1 or H3 on the human virus was swapped for H5.  However, the resulting virus would be H1N1 or H3N1 and would not meet WHO's own definition of a pandemic virus which has to have a non-human sero-type, like H5.

This year the WHO press releases do not talk as much about reassortment, but they talk more about small changes via random mutation.  The small changes are correct, but these changes are happening via recombination, not mutation.  This can be easily seen via analysis of the 2004 sequences at GenBank and inferred by the descriptions of the 2005 sequences offered on press release (the actual changes cannot be seen by the public because the various labs that have sequenced the 2005 isolates have not made the sequences publicly available).  There are currently no 2005 H5N1 sequences at GenBank and the sequencers who have the sequences still haven't figured out how H5N1 evolves.

H5N1 evolution is quite clear when the data at GenBank is analyzed properly.  In 2004 H5N1 exploded throughout Asia.  Most of the isolates were the Z genotype, which defines are particular constellation of H5N1 genes.  The Z genotype was formed via reassortment.  However, there are regional differences between the various Z genotypes.  Thus, the Z genotype is different in Vietnam, Thailand, Indonesia, South Korea, Japan, and various provinces in China.

Some of the most striking differences are seen in isolates from Vietnam and Thailand in 2004.  Although there are markers found only in Vietnam and others found only in Thailand, there are a large number of markers found exclusively in both Vietnam and Thailand.  These makers are not in other H5N1 isolates, but are in other sero-types found in mammals (H1N1, H1N2, H3N2).  Thus, the isolates in Vietnam and Thailand are recombinants that have acquired mammalian polymorphisms and the only reported human H5N1 infections in 2004 were in Vietnam and Thailand.  The amantadine resistant markers are a couple of additional examples.  One marker (at position 31 in the M2 protein) is found in a small number of isolates outside of Vietnam and Thailand (but in 2004 there was only one, which was from Guangdong).  In contrast, Genbank has 13 M2 sequences from Thailand and 16 from Vietnam and all 29 sequences have the amantadine resistance change at position 31.  However, these same 29 isolates have another change at position 26 in M2 and that change is not found in any H5N1 isolate at GenBank.  Thus, these two markers are examples of regional changes for in Z genotype isolates.

The amantadine resistance appears to be present in 2005 isolates also.  Media reports indicated that the first 9 isolates from southern Vietnam this season have the amantadine resistance marker(s).  this is not a surprise because although H5N1 can change rapidly, most changes are stable.  The stability was seen in 2004 when isolates from the beginning of the season were compared to isolates at the end of the season.  There was a familial cluster in Vietnam at the end of July, which was followed by a familial outbreak in Thailand as well as an outbreak at a tiger zoo.  In addition, a number of isolates were obtained from birds in the summer of 2004 in Thailand.  Analysis of these sequences showed that they were very similar to the isolates from Vietnam and Thailand collected at the beginning of 2004.  These sequences were stable because the recombination in 2004 within Vietnam and Thailand did not produce much genetic change because the recombining genomes were very similar.

However, they situation changed this season because migrating birds and or trade brought in new sequences.  Thus the H5N1 in northern Vietnam has acquire the HA cleavage site that had been seen in 2003 and 2004 isolates in China and Japan.  However, although that region of HA had been acquired via recombination, much of the gene had the same features as 2004, and therefore could still cause disease in humans.  The virus in southern Vietnam ad Cambodia was still like the 2004 isolate, killing the vast majority of infected people, but the isolate in northern Vietnam was less lethal, but more easily transmitted.  Since H5N1 had become endemic to Vietnam and Thailand in 2004, there were many targets for recombination in 2005, leading to a marked increase in diversity.

Today Vietnam has acknowledged that these changes have lead to false negatives and the extent of H5N1 infections is highly than previously acknowledged.  The admission in Vietnam suggests that false negatives were also generated in Thailand, since 2005 in Thailand is similar to 2005 isolates in northern Vietnam, where the number of mild human cases has markedly increased. 

Since much of this change has only been recorded on western blots, the dramatic increases have not been reflected in the "official" numbers maintained by WHO.  WHO has issued a press release indicating that these positives were "research" results and required confirmation by HI tests, which are said to not be possible in Vietnam because of a lack of a BLS-3 facility.  The failure to include this data in the official totals, even though the samples were collected in March or early April is yet another example of how scandalously poor the H5N1 surveillance is, even in the country with the highest number of reported H5N1 cases.  H5N1 changes not only affect tests designed to detect the virus, but these changes also affect the clinical presentation, which has been used to exclude testing of a large number of patients in Asia with "mysterious" disease.

Thus, at this point WHO has a very incomplete data base on where H5N1 is and isn't and how evolved the virus has become in 2005.  These two glaring deficiencies will significantly impact efforts to control the 2005 H5N1 flu pandemic.

Media sources

Home | Founder | What's New | In The News | Contact Us

© 2005 Recombinomics.  All rights reserved.