Robert Webster on H5N1 evolution, vaccines 12-11-06

[4-ABBA]

http://www.alertnet.org/thenews/newsdesk/SP3291.htm

Vaccines for all H5N1 flu strains crucial -experts
11 Dec 2006 10:15:15 GMT
Source: Reuters

By Tan Ee Lyn

SINGAPORE, Dec 11 (Reuters) - The H5N1 bird flu virus has undergone many changes since making its first known jump into humans in 1997 and vaccines must be manufactured to fight its major strains, experts said on Monday.

While the virus remains largely a bird disease and does not infect people easily, the scientists at a conference on avian flu and other infectious diseases in Singapore warned against any complacency.

"What's worrying is there were more (human) cases in 2006 than 2004 and 2005. The problem is still with us," Robert Webster of the St Jude Children's Research Hospital in the United States told Reuters on the sidelines of the conference.

"It's (H5N1) continuing to evolve and there are multiple lineages of this virus still out there. What cross-protection is there between these clades (strains) and sub-clades?"

Webster said several H5N1 strains had become widespread and different enough to cause unease among experts, and no one would dare assume that any one vaccine would be able to protect against other H5N1 strains.

"They are sufficiently different so that we're all making vaccine strains against each one of these -- we are making a large number of vaccine strains in case any one of these became dominant," he said.

ENOUGH PROTECTION?

H5N1 has killed 154 people since 2003 and experts fear it might trigger a pandemic if it learns to transmit easily among people.

Derek Smith of Cambridge University in Britain said there have been at least five major changes to the H5N1 virus since it was first discovered in 1959.

These five strains were found in Hong Kong in 1997, Vietnam in 2004, Eurasia-Africa in 2005-2006, Indonesia in 2005 and Anhui province in China in 2005.

"It's not clear what is driving this antigenic evolution," Smith, research associate at Cambridge's zoology department, told the conference.

Several companies around the world are in a race to develop vaccines against the virus, although many experts think they might not confer protection against an eventual pandemic strain, if one should emerge.

Viruses mutate constantly and since vaccines are made based on a selected strain, they might not work as well against other strains.

Some experts at the conference also urged governments in the developing world to be realistic. In the event of a pandemic, there simply would not be vaccines or drugs for poorer nations.

"The world's vaccine production capability is about 350 million doses of flu vaccine per year, so it's going to reach only a very small population of the world," said Roy Anderson of the department of infectious disease epidemiology at Imperial College in London.

More time and resources should be spent instead on researching about, and promoting, simple hygiene measures "that might be beneficial to the majority of the world at very low cost", he said, citing the use of surgical masks, alcohol sprays and regular hand-washing.

 

[AlaskaDenise]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

"It's not clear what is driving this antigenic evolution,"

...this lack of understanding is not reassuring.

...In the event of a pandemic, there simply would not be vaccines or drugs for poorer nations.

"The world's vaccine production capability is about 350 million doses of flu vaccine per year, so it's going to reach only a very small population of the world,"

...So why not put less money into vaccine research and more into vaccine factory construction?
Or take a small amount of our defense budgets!

Also if we need 2 doses to get sufficient immune response (as many tests have shown), then those 350 million become 175 million GLOBALLY!

More time and resources should be spent instead on researching about, and promoting, simple hygiene measures "that might be beneficial to the majority of the world at very low cost", he said, citing the use of surgical masks, alcohol sprays and regular hand-washing.

...sounds like 1918!

They need to read the papers referenced in JJackson's post today about aerosol sizes and masks - only respirators will work.
See post # 6 on thread at http://www.flutrackers.com/forum/showthread.php?t=13995

.

 

[4-ABBA]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

New vaccine production facilities take 2+ years to build from design to completion. Can't remember where I read that--but it's been quoted several times in the press.

 

[sharon sanders]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

I would like to remind everyone that this site is apolitical.

 

[Harpsong]

Webster: Keep Small Birds Out of Chicken Farms

Webster: Keep Small Birds Out of Chicken Farms

Keep small birds out of chicken farms, expert says
11 Dec 2006 11:31:00 GMT
Source: Reuters

SINGAPORE, Dec 11 (Reuters) - Poultry farms must be properly screened and protected from small birds such as sparrows, starlings and pigeons which are capable of passing the H5N1 bird flu virus to chickens, an expert said on Monday.

These small birds are resident in many countries and small numbers of them have been found infected in recent years with the H5N1 virus.

Leading virologist Robert Webster told Reuters his laboratory infected sparrows, starlings and pigeons with strains of the H5N1 virus isolated in Vietnam, Thailand and Hong Kong recently.

His team confirmed the birds shed the virus in their stools and can therefore infect poultry.

"They were infected and shedding the virus in their faeces and from their respiratory tracts. The sparrows died, so they are not as big a threat," Webster said on the sidelines of a conference on avian flu and other infectious diseases in Singapore.

"The bigger threats are the starling and the pigeon. The starling didn't die, but shed plenty of virus," said Webster of the St Jude Children's Research Hospital in the United States.

The virus replicated very well in the starling and less well in the pigeon, he added.

Although all three species did not transmit the virus to their own kind, the fact that the infected starlings and pigeons did not succumb to the virus meant they could be dangerous to poultry.

"This means that you've got to keep these small birds out of chicken houses, too, because they can be infected and they can carry the virus from this chicken house to that chicken house," Webster said.

The H5N1 virus remains largely a bird disease and does not infect people easily. However, it has killed 154 people since 2003 and experts fear it might trigger a pandemic if it learns to transmit easily among people.

Experts have long urged countries such as China and Indonesia to enforce stronger surveillance on poultry and to get rid of the virus in chickens and domestic ducks because they are seen as major agents in transmitting the virus to people.

Most of the 258 people infected with H5N1 since 2003 are believed to have contracted the disease from chickens. They kept chickens in their backyards and many were believed to have handled sick and dead chickens before falling ill themselves.

(Editing by David Fogarty; World Desk Singapore +65 6870 3813)

 

[gsgs]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

but, how can these alternative measures be efficient without vaccine ?

As long as people are naive to the pandemic strain, they are
suspect of infection.
So the plan was to delay the spread until vaccine is available.

Can we reasonably assume that we might be able to stop a pandemic
with handwashing etc. ?
When you don't get it this week then maybe next week or next year.
OK, there is some hope maybe that it's less virulent then.

In previous flu-pandemics, it seems that everyone was eventually
exposed to the virus. Some got disease, some not.
But 20%-40% of people became infected, there are no examples
with 10%,5%,2% being infected, I think.
So can we realistically hope to efficiently reduce the number of infected
people without immunization measures ?

If delaying an infection should work, how long must it be delayed until
you are safe ? 1 year, 3 years, 10 years ?

 

[AlaskaDenise]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

So can we realistically hope to efficiently reduce the number of infected
people without immunization measures ?

If delaying an infection should work, how long must it be delayed until
you are safe ? 1 year, 3 years, 10 years ?

May be answered in MHSC's thread at http://www.flutrackers.com/forum/showthread.php?t=14031

This sounds like one of those many many things we don't fully know about influenza.

.

 

[gsgs]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

AD, research is important, we must know how to best do this,
this is still unclear , else we might make mistakes.

Vaccine production capacity is not a static number.
When there is demand for vaccine and it is well paid,
then there will be additional capacities.

Just, how much are we willing to pay for it ?

 

[JJackson]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

From GSK testimony to Royal Society report (for full text follow link in AD post #2 of this thread)

GSK does not see the development of pandemic influenza vaccines as a commercial activity; it is led by public health need. A combined effort will clearly be required between the EU, WHO, governments and industry to
respond to a pandemic flu situation.

 

[AlaskaDenise]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

........Just, how much are we willing to pay for it ?

Just how much is the cost of a 1918 level pandemic?

Want to venture a cost estimate for higher CFRs?

.

 

[gsgs]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

just, how likely is this ?

 

[GenePalmer]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

Quote:
"It's not clear what is driving this antigenic evolution,"

...this lack of understanding is not reassuring.

...So why not put less money into vaccine research and more into vaccine factory construction?

Hi All,

I'm still a newbie here, and trying to play catch-up, but barely scratching
the surface. Perhaps this is bad netiquette to dredge up an old thread,
but I find it a bit troubling that some are referring to recombination driving
the evolution of flu HA.

While a copy choice mechanism may play a role in genetic diversity, the
intrinisically high error rate of influenza RNA polymerase would satisfy
Occam's Razor just fine. Spinning off mutants at a much higher rate than DNA viruses, flu exists as a quasispecies. Any positive adaptive advantage
one mutation confers on that particular variant results in a shift in proportion
favoring that variant, and it's ability to outgrow others. Conversely, negative
selective pressures, like virus-neutralizing antibodies, shift the quasispecies
population toward escapes mutants that are still viable.

Now apply this selective pressure in poultry who were vaccinated either
with suboptimal doses, or vaccine derived from a strain no longer circulating
in migratory waterfowl as the predominant strain (or both), and you have a classic incubator for hatching the Fujian-like viruses. The clade 2 viruses in Indonesia, likewise, probably represent strains selected for in domestic fowl by hoarding vaccine.

1.2 billion USD has been awarded by the US government to fund both
pandemic vaccine development AND the infrastructure to manufacture
seasonal and pandemic vaccines. We need both. What good is there
to have large scale bioreactors churning out vaccine if the strain most
likely to break has not yet emerged?

Funding a two-pronged approach seems most prudent. Learn to scale
up production in new cell-based substrates using seasonal flu vaccine as
the model, while focusing research efforts with H5N1 on potentially conserved
epitopes that could induce highly cross-protective virus-neutralizing antibody.

In the meantime, Rob is quite right to promote a cultural change among
people who comingle birds in their backyards. Vietnam created a blueprint
but even in the face of diligent implementation, new outbreaks are emerging
in poultry. Still, a scrupulously administered poultry vaccination program
can do much to reduce the "bridge reservoir". Isolating chickens from ducks,
and both from wild waterfowl might, in the end, be the most cost-effective
way to control H5N1 transmission to humans.

If in the event H5N1 ever becomes efficient at human-to-human transmission,
an effective vaccine would be indispensible. Choosing the right one is the
trick.

~gene

 

[wetDirt]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

Quote:
"It's not clear what is driving this antigenic evolution,"

...this lack of understanding is not reassuring.

...So why not put less money into vaccine research and more into vaccine factory construction?

Hi All,

I'm still a newbie here, and trying to play catch-up, but barely scratching
the surface. Perhaps this is bad netiquette to dredge up an old thread,
but I find it a bit troubling that some are referring to recombination driving
the evolution of flu HA.

While a copy choice mechanism may play a role in genetic diversity, the
intrinisically high error rate of influenza RNA polymerase would satisfy
Occam's Razor just fine. Spinning off mutants at a much higher rate than DNA viruses, flu exists as a quasispecies. Any positive adaptive advantage
one mutation confers on that particular variant results in a shift in proportion
favoring that variant, and it's ability to outgrow others. Conversely, negative
selective pressures, like virus-neutralizing antibodies, shift the quasispecies
population toward escapes mutants that are still viable.

There is substantial evidence from the sequence data that the diversity of strains is being driven by multiple infections with slightly different strains of the virus. This is pretty obvious by comparing the patterns of change between the strains. A random mutation model would predict random differences between strains, but the data shows systematic differences. So while random drift is a simple conceptual model, it is not consistent with actual data. So out it goes. We are left with the problem of a minimum 6-month lag between selecting a strain and having the vaccine in hand, but meanwhile, the virus has produced new strains in the interim. Regardless of the mechanism, this problem won't go away. It's an intrinsically hard problem.

 

[JetStuuby]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

Hi,

Gene's comment above seems to indicate that he beleives recombination is NOT the method of influenza evolution.

Isn't the current data clearly showing that the virus is not altering randomly, but is altering in a non-random way?

Sorry, i'm not a geneticist, but i beleive random mutation means just that, random changes at all points of the gene of the virus, yet from all the travel logs i have seen posted here there appears to be 'sequences' or 'regions' (sorry am a layperson in this field so don't know a better term) that are consistently carried throgh the travel log.

My university level understanding of mathematics makes it seem that random mutation would see these 'regions' appear and then dissapear again randomly rather than appearing and then remaining.

I think that if future eqyptian smaples continue to show 'regions' that are adapted to humans remaining in the samples, then this indicates some form of darwinian evolution that simply IS NOT random.

Does this sound logical, or am i miss-understanding the apparently simple and blatant logic of recombination?

THanks to anyone who can backup or dispel my understanding,

Cheers Jet

 

[GenePalmer]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

There is substantial evidence from the sequence data that the diversity of strains is being driven by multiple infections with slightly different strains of the virus. * This is pretty obvious by comparing the patterns of change between the strains. A random mutation model would predict random differences between strains, but the data shows systematic differences. So while random drift is a simple conceptual model, it is not consistent with actual data. So out it goes. We are left with the problem of a minimum 6-month lag between selecting a strain and having the vaccine in hand, but meanwhile, the virus has produced new strains in the interim. Regardless of the mechanism, this problem won't go away. It's an intrinsically hard problem.

* quasispecies: "slightly different strains" all in the exposure population...

my discussion of selective pressures does not imply randomness at all.
you start with a mixed bag because the virus is sloppy; has very little
fidelity. what emerges is the predominant strain which enjoys the best
fit. it's hard to imagine a multiple infections model when it would only
require one infection with a quasispecies to give the same result.
Back to Occam's Razor.

 

[wetDirt]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

* quasispecies: "slightly different strains" all in the exposure population...

my discussion of selective pressures does not imply randomness at all.
you start with a mixed bag because the virus is sloppy; has very little
fidelity. what emerges is the predominant strain which enjoys the best
fit. it's hard to imagine a multiple infections model when it would only
require one infection with a quasispecies to give the same result.
Back to Occam's Razor.

I would invite you to look at some of the sequence data before marrying this hypothesis. As I said, the sequence data are not consistent with a low-fidelity virus. Occam's razor needs to be sharpened occasionally, it gets old and dull when used on really old ideas.

 

[GenePalmer]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

I'm not here to debate or argue. My desk is piled with alignments, so yes,
I've seen the sequence data. Nothing I see is inconsistent with selective pressure on a quasispecies.
see:
http://www.cdc.gov/ncidod/EID/vol12no01/pdfs/05-1024.pdf
http://publish.csiro.au/paper/EA03102.htm
http://www.futuremedicine.com/doi/abs/10.2217/17460794.1.3.255?cookieSet=1&journalCode=fvl
http://www.futuremedicine.com/doi/abs/10.2217/17460794.1.2.243?journalCode=fvl
http://www.bioone.org/perlserv/?request=get-abstract&doi=10.1637%2F6093

and though quite old, still useful:
http://pespmc1.vub.ac.be/QUASIS.html
http://www.microbiology.wustl.edu/dept/fac/huang/ccas/mut/sel.html

Point is, I didn't just pull this <SNIP>. I've studied this for awhile now.:tiphat:

 

[Niko]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

Welcome aboard GenePalmer!

It seems we were posting at the same time. I look forward to reviewing the references you provided.

I would invite you to look at some of the sequence data before marrying this hypothesis. As I said, the sequence data are not consistent with a low-fidelity virus. Occam's razor needs to be sharpened occasionally, it gets old and dull when used on really old ideas.

Since this is not my field and I am unqualified to assess the data, I offer up one of Dr. Niman's commentaries which discusses the preservation of large segments of sequences over years (decades):

Commentary
[FONT=Arial,Helvetica].[/FONT]
Swine Recombinants Destroy Influenza Genetics Basic Tenets

Recombinomics Commentary
March 23, 2006

The recent Canadian swine H1N1 and H1N2 sequences have many examples of recombination as indicated earlier. However, a more detailed analysis of these sequences identifies a series of recombinations that effectively destroys the two basic tenets of influenza genetics which define genetic drift and shift. Genetic drift is said to be due to random mutations introduced by a ?sloppy? polymerase which poor replicates the eight gene segments of influenza. Genetic shift is said to be due to whole genes being swapped when a host is infected with two distinct flu strains. Although transcription errors do become fixed in circulation viruses and the viral genes do reassort during dual infections, the gene changes that are seen on an annual basis are largely due to recombination. The recombination can produce small changes such as single nucleotide changes (creating drift) or can swap large portions of individual genes (creating shift).

The recombination was readily seen in the recent swine isolates because full gene sequences of all eight segments were made public, and the recently released sequences not only had portions of genes that exactly matched prior isolates, but they also have portions of genes that exactly matched other recent isolates. These relationships were particularly striking and obvious in the PB2. Some of the acquisitions of portions of previously described isolates was delineated earlier. However, most of the portions of genes that did not match earlier isolates did match the current isolates.

One example was the parings in the PB2 gene of 11112 and 23866. The two genes matched each other in the first and last third of the sequence. However, most of the middle third of 23866 matched the 1977 sequence from the swine isolate from Tennessee, while the middle third of 11112 match the 1998 swine isolate from North Carolina.

Another example was the parings of the PB2 gene of 56626 and 53518. The genes matched each other for the first 550 BP, but the remainder of the 53518 sequence matched the 2002 sequence from Taiwan. In 56626 the middle 1000 BP matched the same region in 57561 and the region that was an exact match between the two recent isolates also contained the core region that matched the 1977 Tennessee sequence.

The third example was the pairings between 48235 and 55383. The first 1850 BP in the two sequences were exact matches, but this region also included over 1600 BP that also were an exact match of the 1977 Tennessee sequence.

Thus, these pairing clearly showed that the PB2 genes were generated by a series of recombinations involving large portions of the genes which constituted genetic shifts. Moreover, the identity of large positions of genes between isolates from 2003 or 2004 and 1977 indicated that the polymerase could copy the sequences with a high degree of fidelity, producing an exact match between isolates that were at least 26 years apart.

As noted earlier, the same 1977 isolate from Tennessee contributes large portions of the PA genes found in six of the seven isolates. The remaining avian genes have additional examples of easily identifiable recombination.

The swine sequences evolved slowly, preserving the identity of sequences acquired earlier via recombination. These large acquisitions can also be found in more rapidly evolving avian sequences, although the multiple sets of recombination reduces the size of the acquisitions. Moreover, in many instances the region that was expected to contain newly acquired sequences was missing from the sequence database. Full sequences of these isolates includeing H5N1 bird flu, would produce additional examples of recombination.

There also appear to be more instances of acquisitions of smaller regions which frequently have a high degree of identity, so the acquired new sequence may be limited to a single nucleotide change. The recombinant nature of these smaller differences can be traced to the same parent contributing new genetic information to multiple locations with a gene segment or acquisitions of new genetic information in multiple segments.

These two types of recombination account for the vast majority of the year to year changes in related influenza genes.

http://www.recombinomics.com/News/03230601/Swine_Recombinants_Destroy_Tenets.html

A portion of another commentary:

H5N1 Random Mutations Are Not Random Or Recent Mutations

Recombinomics Commentary
March 21, 2006
http://www.recombinomics.com/News/03...tions_Not.html


As the number of bird flu sequences increase it becomes more apparent that the gentic drift of H5N1 is not due to random mutations, but is due to recombination. The widely held belief that the drifting is due to copying errors becomes less and less tenable as new sequences are made public.

Much attention has been focused on the Qinghai strain of H5N1 because it has become resident in long range migratory birds and is rapidly spreading worldwide. Most recent confirmed cases are in Europe, the Middle East, and Africa, but the same strain is likely involved in Asian countries such as India, Pakistan, and Afghanistan also.

The Qinghai strain was initial isolated at the Qinghai Lake nature reserve in central China. The strain was readily distinguished from other H5N1 isolates from Asia. Its HA cleavage site was GERRRKKR instead of the more common RERRRKKR found in Asia. In addition, the PB2 polymorphism E627K was in all 16 isolates. Prior to Qinghai Lake, this changed had never been reported in H5N1 isolated from birds. It was in all H1, H2, H3 human isolates or H5N1 from mammals such as humans, wild cats, or brains of experimental mice, but not from birds. Since Qinghai Lake, all H5N1 Qinghai strain PB2 sequences containing the 627 position have had the E627K polymorphism. Moreover, all of the Qinghai strain isolates are over 98% homologous to the Qinghai strain and possess a number of Qinghai specific polymorphisms in all 8 gene sequences.

http://www.recombinomics.com/News/03210601/H5N1_Random_Mutations_Not.html

Both of the above may be found on Dr. Niman's site (links provided above) and are also posted here at FT in the Recombination forum found at: http://www.flutrackers.com/forum/forumdisplay.php?f=9

I have no "pony in this show" and would love to see a respectful and thorough discussion of the mechanism(s) by which avian influenza virus change and develop new strains. Following this topic for 2 years now, I have yet to see such a discussion which doesn't devolve into a slap fest.

 

[GenePalmer]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

Thanks for the warm welcome, Niko. And thanks also for the references
to recombo. I'm not an expert, though I have followed debates concerning
recombination vs. selective pressures for some time now. Recombination
can be a powerful tool to drive evolution, but all too often is the hand
waving some fall back on to "prove" an evolutionary mechanism or even
explain an peculiar isolate. Remember Henry Niman founded a company
called "Recombinomics". I'm sure promoting this mechanism serves his
agenda. But where is the proof? In the sequence data? Both hypotheses
can explain these data.

The fact that 1918 isolates of H1N1 separated in space and time were
99% identical at the nucleotide level certainly waters down the notion of
a sloppy polymerase. On the otherhand, if what one means by invoking
this error proneness is to suggest the virus has the potential to spin
off many variants in the course of a productive infection, one can
argue that a quasispecies mechanism can provide the substrate for
antigenic drift.

It is this population of variants that get swamped out by the fittest strain
in the population. Others are not isolated by RT/PCR cloning as cDNA, or even represented in PCR sequencing at a signal level high enough to be
detected.

The process is not random. The variants likely are (with some tansversion/
transition bias). What emerges appears to be one strain. I assure you it
is not. It is predominantly one variant that enjoyed the best fit for that
host. Let it jump from migratory waterfowl into chickens, then into man
and you can see certain hotspots (ie; N186S and Q196R in H5 HA).

That these substitutions arose from recombining HA sequences from another
coinfecting strain of flu vs. the single nucleotide point mutations required
to provide the same result can be debated. The salient point I would like to propose is that selective pressure on a population of variants all within the same productive infection appears to have been overlooked in this thread. Evolution favors the fittest. I'd like to know which mechanism is really
operable.

 

[4-ABBA]

Re: Robert Webster on H5N1 evolution, vaccines 12-11-06

I'd like to know which mechanism is really
operable.

Why does it have to be either / or? Is it possible they both are?