Friday, March 24, 2006

Sugar, Sugar ...

As of March 24th, over 186 people have tested positive for the avian flu virus (H5N1) and 105 are dead (1).

This is a nasty kind of flu.

Luckily, most of the cases, so far, have been in our avian friends and their human companions. Health professionals all over the world, though, are warily watching web sites and looking for signs that people are catching flu from other people. Everyone wants to know the point when it's no longer strictly for the birds.

But for the moment, we're okay, it seems like the avian flu doesn't spread easily from one human to another.

And now we know why.

Two reports this week, in Nature from Shinya, et. al.(2), and in Science from van Niel, et. al.(3) provide the answer.

Avian flu, it seems, likes to stick to a certain kind of sugar. That modified sugar, is a sialic acid with an alpha 2,3 link to galactose, and is joined through other sugar residues to the surface of special kinds of cells. At one time, flu researchers didn't think humans had these specific kinds of cell-surface sugars. But both of research groups found extra-special sugar coating on the surface of cells, deep in the lungs, and confirmed that these were the cells that got infected.

Van Reil and colleagues found this result by mixing inactivated viral particles with tissue samples. They used fluorescent antibodies to see where the virus stuck and which kinds of cells it liked.

Shinya, et. al. stained infected tissues with lectins to see what kinds of sugars could be found on different cells. They also stained infected tissues (epithelial and alveolar) to look for viral particles.

Since the cells that get infected are located deep in the lungs, any new viral particles produced through an infection have to travel a long distance in order to get out of the mouth and infect someone new. This would make it harder for the virus to infect a new human because it doesn't escape the body through a simple cough or casual sneeze.

How is this different from the current human flu?

The influenza strains that are doing most of the damage, in humans this year, bind to a different kind of sugar than the avian strain. The human flu sticks to a sugar with an alpha 2,6 link to galactose; unfortunately for us, this sugar is found on cells in the nose and upper respiratory tract. This makes the human flu more infectious (for us) since it can travel long distances with a good strong sneeze.

Click the drawing to make it bigger. But be warned, I took some liberties here. Influenza is usually drawn like it comes from outer space. Typical images show lots of spikes to represent the hemagglutinin and neuraminidase molecules on the surface of the particle. Also - the cell and the virus are not drawn to scale.

What does this mean? Are we safe?

No one knows. Some experiments earlier this month, by Stevens, et. al.(4) found that it only took two mutations, in the right positions, to change the specificity of a hemagglutinin from one kind of sugar to another. No one knows how many mutations would be needed to change the specificity of the H5 hemagglutinin. Neither do we know much about the probability that this will happen (although I'm sure this is something that could and should be examined by mathematical modeling).

It could happen.


1. World Health Organization. WHO Cumulative Number of Confirmed Human Cases of Avian Influenza A/(H5N1) Reported to WHO. ( (accessed March 24, 2006).

2. Shinya K, Ebina M, Yamada S, Ono M, Kasai N, Kawaoka Y. 2006. Avian flu: influenza virus receptors in the human airway. Nature. Mar 23;440(7083):435-6.

3. van Riel, D., Munster, V., de Wit, E., Rimmelzwaan, G., Fouchier, R., Osterhaus, E., and T. Kuiken. 2006. H5N1 Virus Attachment to Lower Respiratory Tract. Sciencexpress.

4. Stevens J, Blixt O, Tumpey TM, Taubenberger JK, Paulson JC, Wilson IA. 2006. Structure and Receptor Specificity of the Hemagglutinin from an H5N1 Influenza Virus. Science. Mar 20; [Epub ahead of print]


technorati tags: , , ,


Blogger Paul Decelles said...

Am I correct that the H5N1 virus could recombine with human influenza viruses? That would perhaps provide another way for the virus to pick up binding ability to the other sugar.

On the other hand, perhaps if the virus is able to attach to cells in the upper respiratory tract, maybe the result would be a less virulent virus.

8:33 PM  
Blogger Sandra Porter said...

We just got a new issue of Science in the mail yesterday and it had a paper by Obenaur et. al, Science 311:1576-1580. They reported sequencing 2196 genomes of avian influenza virus. And they didn't see any sign of homologous recombination. Their conclusion is that, if recombination happens with influenza virus, it's rare.

They did see evidence of reassortment (aka antigenic shift) and antigenic drift. Reassortment occurs because each virion has 8 pieces of RNA, that can be exchanged whenever a cell is infected with different strains of virus. They found that the surface glycoproteins (hemagglutinin and neuraminidase) seem to be exchanged most frequently. These proteins also had more variability in amino acid sequence than any of the other proteins (antigenic drift).

They also proposed that the ability of the avian virus to kill human cells might be related to the NS1 protein and that the "avian" NS1 protein is more harmful to human cells than the version from the "human" flu.

Time will tell.

12:07 PM  
Blogger Nelumbo said...

interesting post!

9:31 AM  
Anonymous realityscience said...

its an intresting post just read a recent article about H5N1 about why it doesnot spread that frequent in humans.

11:51 AM  
Blogger Sandra Porter said...


I want to add something about transmissibility that I read at Effect Measure. They pointed out that we need much more data on viral transmission before we can conclude that the disease is hard for humans to catch.

12:00 PM  
Blogger wolf21m said...

Well written in a way that was easy to understand. Thank you.

6:02 AM  

Post a Comment

Links to this post:

Create a Link

<< Home