Friday, May 1, 2009

Exclusive Interview: CDC Head Virus Sleuth

April 29, 2009

Virologist Ruben Donis, chief of the molecular virology and vaccines branch at the U.S. Centers for Disease Control and Prevention, spoke with ScienceInsider at length last night about the swine flu virus causing the current outbreak. CDC’s early analyses raise several provocative possibilities. The stage appears to have been set for this human outbreak by an outbreak over the past decade of flu viruses in swine that combine strains from several species. The first infected human may not even have been in North America, let alone Mexico. Patient samples from Mexico taken over the past several months reveal that this swine flu clearly exploded in late March, suggesting that it was not rapidly spreading in that country, undetected, for very long.

Donis discussed the genetics of the virus—the clues in this mystery—in detail. These include several of its eight genes, which code for surface proteins hemagglutinin (H) and neuraminidase (N), the matrix that surrounds the nucleus, the nucleoprotein itself, and three polymerase enzymes called PA, PB1, and PB2.

—Jon Cohen

Q: What do you know about this swine flu virus?

R.D.: We know it’s quite similar to viruses that were circulating in the United States and are still circulating in the United States and that are self-limiting, and they usually only are found in Midwestern states where there is swine farming. There’s only one well-documented case where the infection spread from one human to another. What we know is that it is not common that there is sustained transmission in people.

Q: Christopher Olsen published a paper that looked at the literature back to 1958 and only found 50 cases of humans infected with a swine influenza.

R.D.: If we have two documented cases a year, maybe that’s just the tip of iceberg. Maybe there are 10 times more or 50 times more. But still, it’s only swine to human, and it stops there.

Q: Have you completely sequenced this virus?

R.D.: Yes, 2 weeks ago. Very soon after we received specimens from California and Texas. Hemagglutinin, neuraminidase, and matrix, the three genes that have the most public health interest, were sequenced, and then the whole genome was completed.

Q: How large is the genome?

R.D.: It’s 14 kilobases.

Q: It’s only a little larger than HIV. You could sequence it in hours.

R.D.: Yes. It’s tiny.

Q: How close is it to the closest strain you know of?

R.D.: Depends which genes. You have similarities of about 94% in the hemaggluttinin [H] to the nearest strain we know.

Q: Is it of swine origin?

R.D.: Definitely. It’s almost equidistant to swine viruses from the United States and Eurasia. And it’s a lonely branch there. It doesn’t have any close relatives.

Q: How about the neuraminidase gene?

R.D.: It has close relatives in Asia. It’s also swine.

Q: The matrix gene?

R.D.: The same as neuraminidase.

Q: So where are avian and human sequences?

R.D.: We have to step back [to] 10 years ago. In 1998, actually, Chris Olsen is one of the first that saw it, and we saw the same in a virus from Nebraska and Richard Webby and Robert Webster in Memphis saw it, too. There were unprecedented outbreaks of influenza in the swine population. It was an H3.

Q: They were dying from it?

R.D.: No. It was not very severe in healthy pigs. Everyone was very curious about these H3 viruses. Since 1918, normally it’s only H1N1 in swine. Then all of a sudden there’s H3N2 in swine in the Midwestern U.S. When people analyzed what was inside those viruses, they realized there were three different things.

The PB1 gene, that was human. H3 and N2 also were human. The PA and PB2, the two polymerase genes, were of avian flu. The rest were typical North American swine viruses. Those strains were the so-called triple reassortants.

Q: We always hear of the pig as a mixing vessel combining human, avian, and pig influenzas. Why didn’t you regularly see reassortants?

R.D.: Good questions. These questions have no answer. There is an explanation somewhere.

The reality is good molecular surveillance in the pigs started in the 1970s. So if there were strains that were not very dominant between the 1930s and the ’70s, we wouldn’t have detected them. This triple reassortant was very successful and took over and dominated the picture—to the point where the classical H1N1 was almost extinct.

Q: Why were the first triple reassortants more fit?

R.D.: They were H3s, and H3 is a different subtype, so there was no immunity in the pigs. It was probably that they had new polymerase genes, too.

Q: How does it tie to the current outbreak?

R.D.: Where does all this talk about avian and human genes come from? I was describing a fully swine virus. For [the] last 10 years, this has been a fully swine virus. Can you tell I have an accent? I’m a U.S. citizen but I have the roots in Argentina. It’s like me. I’ve been in the U.S. since 1980. I’m a U.S. citizen but I have an accent.

Q: It’s not as though human and avian just got there in this strain.

R.D.: It’s part of the swine virus.

Q: What’s the newest part of this strain?

R.D.: Neuraminidase and the matrix are the newest to be seen in North America. They were not part of the team—I talk about flu virus as teams of genes. There are eight players. They have these two new players from Asia.

Q: It suggests a mixing of pigs from North America and Asia.

R.D.: One little detail we haven’t discussed is [that] these Midwestern viruses were exported to Asia. Korea and many countries import from the U.S. Swine flu is economically not such a big deal that many countries don’t check for it.

Q: How do you get Europe in there?

R.D.: There are some parts of the puzzle I don’t have the answer to. The genetic lineages of Asia and Europe mix quite a bit.

Q: How does the pig get back here?

R.D.: Who said it was a pig that came from Asia? Did I say that? It could be a person.

Q: So the origin might trace back to Europe or Asia.

R.D.: I didn’t say that. I don’t want to point the finger at anyone.

Q: It does suggest that mixing didn’t happen in Mexico.

R.D.: Probably not. The amazing thing is the hemagglutinins we are seeing in this strain are a lonely branch that have been evolving somewhere and we didn’t know about it.

Q. Can you look at sequence to see what makes it transmit human-to-human?

R.D.: If I could, I would be the chief of the CDC or NIH [National Institutes of Health].

Q: You don’t want either of those jobs.

R.D.: [Laughter.] I don’t think we can do it in silico.

Q: Do you know the percent difference of the entire sequence to the older triple reassortant that dominated?

R.D.: We have [a] 6% or higher percentage difference in neuraminidases. You have multiple amino acids that differ. And single amino acid changes can change receptor specificity. When you have so many changes, you don’t know which ones are responsible.

Q: How about pathogenesis? Can you tease that out in vitro?

R.D.: One traditional approach is to take advantage of viral modules that allow you to assemble different teams, to make reassortants that take a virus say from North America that doesn’t transmit, and you swap one gene from the virus that does transmit. If the hypothesis is that hemagglutinin is responsible, you put in the background of the genes from the old virus. You need an animal model, usually the ferret.

Q: Have you been able to compare isolates from Mexico and the United States?

R.D.: Yes, they are very, very similar. Many genes are identical. In the eight or nine viruses we’ve sequenced, there is nothing different.

Q: Have you compared someone who died with someone who had a mild case?

R.D.: Those data are still slippery. We don’t have good case data. You get age and sex—very limited information. That’s a problem. In the set of samples we know one case was fatal, but we don’t know which one it is.

Q: Have you been to Mexico yet to study this outbreak?

R.D.: No. CDC sent a group of scientist and epidemiologists, laboratory folks that went over there to set up diagnostic labs. One is in Yucatán; one in is in D.F. [Distrito Federal, which includes Mexico City].

Q: What do you think about the pig farm in Veracruz?

R.D.: I don’t know the details. They said they had a huge operation and the workers were not getting sick; that’s what the company claims. The only suspicious thing in that story is this is the largest farm in Mexico. The fact that the index case also is from the area makes it interesting.

Q: Do large farms have more swine flu?

R.D.: Not really. Even folks who have 50 pigs have to buy feed and supply from vendors that go from farm to farm, and they don’t wash their boots or whatever. Usually the virus is transmitted very effectively.

Q: Is there anything I didn’t ask you that I should have?

R.D.: We all pray this remains sensitive to antivirals. We all hope that vaccines will be developed. The virus doesn’t grow very well in eggs. We hope the virus will improve [the] ability to grow in eggs so we can produce [a] vaccine very quickly so these secondary and tertiary cases can be controlled. In some countries there’s good surveillance, but in others, who knows.

Q: What do you think of this outbreak?

R.D.: This is the first one I’ve seen firsthand as a virologist. The avian influenza outbreak is not comparable because this is unfolding so quickly. This reminds me of SARS. With avian there’s very little transmission. And even with SARS, transmission was far less.

Q: Does this one scare you?

R.D.: I saw figures that do scare you. We’ve received 300 samples from Mexico, and these cover the span of February, March, and April. And you look at flu A, traditionally it’s A/H1 or A/H3 or it's B up until the end of March. There are two or three cases up to [the] last days of March that are swine. Then in April they skyrocket. So all the cases in the D.F. areas, where most samples came from, it really transmits very efficiently.

Q: What is the date of first sample?

R.D.: I think it’s the end of March, the first positive specimen.

Q: Did Mexico react quickly enough?

R.D.: They didn’t know. They probably thought it was regular flu.

Q: Flu is a seasonal disease that peaks in winter. Maybe this will end in the United States with the end of the flu season.

R.D.: We’re in a good position. The folks in Buenos Aires are in trouble. They’re entering winter now.

No comments: