Saturday, June 06, 2009

What Do You Get When You Cross Bird Flu With H3N2?

 

# 3303

 

 

This, of course, is one of the big questions scientists have wrestled with over the past few years. 

 

What happens if you take a virulent flu that doesn’t transmit readily between humans (H5N1) and cross it with a mild virus that does transmit easily (H3N2)?

 

We know that two (or more) flu viruses are capable of infecting the same host, and that sometimes dissimilar flu strains can swap genetic material (reassort) and produce a new offspring.

 

We know this because we can see the result.  Every once in awhile we see a new flu strain (like the novel H1N1) show up that came from this reassortment process.

 

But what we don’t know is how easily or often this occurs, and whether the H5N1 bird flu virus is a good candidate for reassortment.

 

 

Today, we’ve a fascinating study that looks at exactly those questions which was published in JVI (Journal of Virology). I’ve taken the liberty to reparagraph the abstract to make it easier to read online.

 

First, the abstract (hat tip to Anne for posting it at Flutrackers), followed by some discussion.

 

 

Reassortment between avian H5N1 and human H3N2 influenza viruses in ferrets: a public health risk assessment

 

Sara Jackson, Neal Van Hoeven, Li-Mei Chen, Taronna R. Maines, Nancy J. Cox, Jacqueline M. Katz, and Ruben O. Donis*

 

Influenza Divison, Centers for Disease Control and Prevention, Atlanta, GA

 

* To whom correspondence should be addressed. Email: rvd6@cdc.gov.

 

   Abstract

This study investigated whether transmissible H5 subtype human-avian reassortant viruses could be generated in vivo.

 

To this end, ferrets were co-infected with recent avian H5N1 (A/Thailand/16/04) and human H3N2 (A/Wyoming/3/03) viruses.

 

Genotype analyses of plaque-purified viruses from nasal secretions of co-infected ferrets revealed that approximately 9% of recovered viruses contained genes from both progenitor viruses.

 

H5 and H3 subtype viruses, including reassortants, were found in airways extending towards and in the upper respiratory tract of ferrets.

 

However, only parental H5N1 genotype viruses were found in lung tissue.

 

Approximately 34% of the recovered reassortant viruses possessed the H5 HA gene, with 5 unique H5 subtypes recovered.

 

These H5 reassortants were selected for further studies to examine their growth and transmissibility characteristics. Five H5 viruses with representative reassortant genotypes showed reduced titers in nasal secretions of infected ferrets as compared to the parental H5N1 virus.

 

No transmission by direct contact between infected and naïve ferrets was observed.

 

These studies indicate that reassortment between H5N1 avian influenza and H3N2 human viruses occurred readily in vivo and furthermore that reassortment between these two viral subtypes is likely to occur in ferret upper airways.

 

Given the relatively high incidence of reassortant viruses from tissues of the ferret upper airway, it is reasonable to conclude that continued exposure of humans and animals to H5N1 alongside seasonal influenza viruses increases the risk of generating H5 subtype reassortant viruses that may be shed from upper airway secretions.

 

 

 

Outside of a controlled laboratory experiment, such as this one, the only time we generally discover that two flu viruses have reassorted, and produced an offspring, is when that offspring turns out to be biologically `fit’ and easily transmitted.

 

Fortunately, that is a pretty rare occurrence.

 

Presumably, reassortments that produce evolutionary dead-ends occur far more frequently.  But once again, we don’t know how often that might happen.

 

This experiment showed surprisingly high percentage of reassorted `offspring’ viruses, with 9% of the viruses recovered from the co-infected ferrets having genetic material from both parental strains.

 

 

The (limited) good news here is that immunologically naïve ferrets, placed in direct contact with these infected animals, failed to contract the virus. 

 

 

In other words, the offspring virus was biologically `fit’ enough to replicate in the upper airway (albeit at reduced titers), but failed the transmissibility test.   And interestingly, these progeny strains failed to infect lung tissue.

 

This time.


Influenza viruses are made up of thousands of amino acids, and we know that the difference in virulence, and transmissibility between viruses can come down to a handful of amino acid substitutions.

 

These individual amino acid changes can occur through a process called recombination, or simply from replication errors while inside a cell.   

 

So while it is certainly comforting to find that this experiment didn’t immediately produce a `Frankenvirus’, that doesn’t mean that a reassortment down the line, in ferrets or perhaps another host, couldn’t.

 

It remains possible that if given enough opportunities (co-infected hosts), and enough replication cycles, that the H5N1/H3N2  (or other) combination could someday produce a biologically fit, and easily transmitted virus.

 

And nature has a large laboratory, infinite patience, and a lot of flu strain combinations to play with.