Big Think

Paul Hoffman: Our next presenter is Michael Worobey. He’s an ecologist and an evolutionary biologist at the University of Arizona. He is part of a team of researchers that have a very important paper published just last month in “Nature,” about the evolution of the swine flu virus. And what he basically found was that it has circulated undetected in swine for perhaps up to a decade. Michael Worobey.

Michael Worobey: Thanks, Paul.

I’m going to talk about the emergence of the swine flu epidemic. But I thought first I put it in the bigger context of the emergence of influenza virus in general.

So here is the kind of tool that I used. It’s an evolutionary tree and all I want you to notice is for this H1 variant here, there is a nice human lineage, the Spanish flu is on that. There is a swine lineage, and there is a bird lineage. And these are the major players in the emergence of human influenza.

And most of the genes, as Peter [Palese] said, of this new variant come from pigs.

So here is a picture from the paper that we published. And just in case this doesn’t mean much to you, all these lines and dots, I want to take you through what we found here.

So coming from British Columbia, I’d like to use the saguaro as the teaching tool. I actually work in Arizona now, so it makes some sense.

With viruses, they evolve so quickly that you can actually see evolution happen, not just on the time frame of years, but in this instance, in the time frame of weeks. And this was a bit of surprise with just a few weeks of sequences that were produced as this pandemic was unfolding, we were able to calibrate the molecular clock.

So let me just use this analogy, if you want to find out how old your saguaro was, you might not be able to do it just by counting tree rings or something. But if you have a photo from 1999 of this cactus in your yard, and you were able to compare it to how much it has grown in the ten years since that time. You start to have a grasp on how quickly this thing grows. And you can make some estimates from the different branches of the percent of that centimeter per year of growth rate, and that can allow you to work back to when the thing originated.

So we do the same thing with viruses. If you have a small observation window like this, you can often make really robust differences about things much deeper in time. Okay now as an analogy, I just want to show you an evolution tree that you are more probably familiar with here. So we do the same sort of things for primates, and if you did an evolutionary tree of primates, you will have a human branch on the tree, or more closely related species, or chimpanzees and bonobos, and then you have the more distantly related species here.

And you can place the date on the most recent common ancestors of all humans and that might be around a hundred years ago. You could also ask, what’s the most recent common ancestor of humans? And whatever is most closely related to them, and that’s all the way back here, around 5 million years ago. So for this swine origin flu pandemic we did exactly the same thing.

So two questions: What is the timing on that ancestor of the virus once it jumped into humans? So when that jumped take place and if you go gene by gene and asked what’s the most closely related species for each gene? It’s actually a pig virus. So we also ask for each gene, when that broke off from the rest of the sample of pig viruses. Okay, so now we’re back to this, and what you see for each gene is the human sequences go less quite recently in time, but nowhere near the March and April [2009] timeline that this thing first emerged, to our knowledge. And for each of the genes, the most closely related sequences are from pigs. Now Peter [Palese] mentioned that some of them ultimately are of avian origin, and some are human origin. What you see when you go gene by gene is that, even for the ones that are ultimately an avian or human origin, they traveled through pigs. And in each case, we find that there’s been a timeline of about 10 years where this virus has been in pigs but has gone totally unnoticed--which says something about our surveillance.

And for the human outbreak, what you see is a timeline of about somewhere around October to December of last year [2008] that the virus was probably in humans. So it took several months to actually spread to the point where the tip of the iceberg became apparent to us.

And here is a rather complicated looking slide that shows while you have genes from an avian pool, and from a human pool, each one of those actually travel through a pig before it got into this new swine origin influenza virus.

One other interesting detail when you look gene by gene and Peter showed a nice picture of the individual segments. So these are actual physically separated chromosomes. You can actually place a date for each one of those as to when it got to humans. When you do that, and you line it up against the answer you get if you take all of the genes together, you get a little bit of a discrepancy. And this discrepancy, it turns out, you get an earlier date with each gene than you do with the entire genome. And we think what’s going on there is once this virus jumps into humans, it actually started evolving at a more rapid phase, and it looks like it goes at about 1.5 times the rate that it evolves in pigs.

And there is some interesting reason why that might be. If you take that into account, and do a little correction for the rate of change, these gene by gene estimates actually starts to overlap the entire genome estimate, and they kind of converge on this timeline of the late autumn, possibly into the very first few weeks of January [2009] that this virus in humans is spreading.

So aside from the actual findings, I just want to spend a couple of moments talking about how this happens. So when the sequences first came out, they were published by places like NCBI [National Center for Biotechnology Information] and GISAID [Global Initiative on Sharing Avian Influenza Data]. And what was great about this situation was that information was made publicly available by pretty much all the parties in real time. So as soon as sequences are available, I started analyzing them, I started firing off emails to colleagues around the world who are doing the same sort of thing. And we set up a Wiki where we can all post our results and critic them, and people that are in competing groups worked together around the clock. We have some people n Arizona, people in the UK, people in Hong Kong.

So for a period of couple of weeks, we were working 16 hours a day and when you fell asleep, you knew that when you woke up in the morning, the research would be advanced another 8 hours, and pretty quickly came to grips with the timeline of this thing.

That kind of open access mentality flowed right through the publication in “Nature.” So “Nature” published this even though we’ve had some press coverage. They also made the paper open access so anyone can download it and look at it.

And I think it’s probably time to open up to the rest of the discussion and I’ll leave the rest of it for them. Thank you very much.

Recorded on: July 14, 2009.