Christian Drosten Credit: Christian Drosten

Christian Drosten is head of the Institute of Virology at the University of Bonn Medical Centre, one of Germany's largest medical virology centres. But he is best known for winning the 2003 race to develop the first diagnostic test for the then-mysterious virus that causes Severe Acute Respiratory Syndrome (SARS). He spoke to Nature about how he has managed to develop a diagnostic test for the H1N1 swine flu even faster, thanks to his experience with SARS.

When did you first learn about the threat of swine flu?

I was at my desk at about 11am on Friday 24 April when the phone rang. It was Stephan Becker, head of virology at the University of Marburg in Germany. He had heard about swine flu from colleagues in America. Both of us had somehow overlooked the first report on ProMed-Mail [the global rapid-alert information webpage for infectious disease outbreaks] — there are so many small reports on that site, it is paradoxically easy for important issues to be missed. It was already being transmitted fast between people, so we knew it was something big. I thought, "Oh no, not now".

What did you do then?

We didn't know when the disease might hit Germany, but we didn't want to find ourselves taken by surprise. So we decided not to wait for test formulations from any authorities, but to try to put together a diagnostic test ourselves — even before there were any suspected cases here. We did not have any physical virus to work with, but the World Health Organization had already made its genome sequence public. I had my experience with the SARS virus, so I knew what had to be done. That experience had also allowed me to make close contacts with other scientists in Germany, and I called on that network of friends on Friday.

So how did you get your materials?

The Institute of Virology in Marburg, which has been working with flu strains from different species for decades, had numerous stored swine flu stocks in its freezer, all of which had been sequenced. It was just a matter of comparing sequences. They found one where the H [haemagluttinin] subunit was virtually identical to the H subunit of the virus now circulating in humans. That's the subunit of greatest importance for diagnosis. Within a day, the Marburg scientists had thawed their samples and extracted the RNA that we needed to make our test.

How do you make such tests?

In general, you need to generate primers to segments of the virus that are unique. It is easy enough to identify these segments using proprietary software to compare sequences. Then you synthesize nucleotide sequences that correspond to these segments. These are called primers, which you use to amplify genetic material from the pathogen if it is present in the test sample.

On Saturday, Marcus Panning at the University of Freiburg identified which primers were needed (while I went to a wedding!). Olfert Landt of the Berlin company TIB Molbiol made the primers physically on Sunday. This part was critical — it is not so easy to get primers physically made to short order, especially over a weekend. I was lucky to have such a good contact in Olfert, again thanks to our work together in the SARS days.

By Monday morning we had everything in place, and we were able to confirm that the test worked in each of our three academic locations — Bonn, Marburg and Freiburg. During Monday we also sent the test to the Bernhard Nocht Institute in Hamburg, where I had worked during the SARS outbreak, and to Hamburg's public health laboratory. As it happened, one of the first suspected case in Germany was admitted to a Hamburg hospital the very next day. Both Hamburg laboratories were able to use out test in parallel straight away. The had the test confirmed by the national reference centre for influenza, the Robert Koch Institute in Berlin.

Can you deal with the number of samples the hospitals in your location are generating?

The numbers we are seeing increase daily. We are getting 10 or 20 at the moment but we expect this to increase explosively. Soon we expect we will have to bring in additional technicians from other departments to cope. We'll have to train them up in this type of testing, but there won't be a problem. Our lab has already been on the TV news, and everyone is keen and curious to help.

Why is it important to do so much testing if the strain is mild?

Because we remember the 1918 outbreak. When it first started circulating in humans, it was not very pathogenic and so it wasn't taken very seriously. But as the virus optimized itself for life in its new host through mutation, it became much more virulent — and, as we all know, killed 20 million or so people around the globe. This could happen again — no one wants to take any big risks.

What tests are other hospitals using in Germany?

By now, many are using our test. Word got around quickly. We have also had inquiries from other countries, so we will also be supplying the test more widely.

Are you planning to develop a new test as the virus mutates?

It is too early to say if this will be necessary — it depends on how exactly it might mutate. But it is actually unlikely to be a reason to change the test later on. In fact, we are already working on a second-generation test that is easier to use for high-throughput screening. We expect to have to test many more samples soon, and we need a test that our technicians can handle with ease.

How does the response to swine flu required of scientists compare to that to SARS and bird flu?

It was much faster this time. Even as we were watching the story break on television news, the virus was already in Germany. We have very good pandemic planning in Germany, but the devil is going to show up in every one of the details. This is going to be a big challenge for people like us who are guiding large hospitals.

Has the response been the same in other countries?

Probably, but to be honest, no one has had the time to exchange worries or complaints in the past few days.

For Nature 's ongoing coverage of the H1N1 outbreak, see www.nature.com/swineflu.