Introduction:

In March of 1997 the influenza research community was startled by the appearance of a remarkable 3-page report� in the journal Science. In the paper Dr. Jeffery Taubenberger and colleagues, of the Armed Forces Institute of Pathology, announced that they had recovered and analyzed fragments of the RNA genome of the virus responsible for the infamous 1918 Spanish Flu pandemic.

The shattered genes of the virus were retrieved from lung-tissue samples fixed by formalin, preserved in paraffin wax, and stored for the last 80 years at the National Tissue Repository, located within the Walter Reed Army Medical Center, in Maryland. The warehouse contains autopsy specimens dating back to the time of the Civil War.

Because of RNA's extreme delicacy, the team's successful extraction and sequencing of the genetic material has both delighted and tantalized those seeking to understand the mysteries of this historically and epidemiologically important influenza strain. Full sequencing of the 1918 genome is expected to take years, although one key gene (coding for the surface protein hemagglutinin) has already been completely pieced together.

Dr. Taubenberger was kind enough to review the manuscript for "Ninth Day of Creation" (a novel in which the 1918 strain is resurrected) and offer useful suggestions for its improvement. The following interview with him about his group's work was carried out through e-mail in May of 1998.

*      *       *      *       *      *      *

Interview:

CRANE: "You've indicated to me that Ninth Day of Creation did a reasonably good job of holding your attention. If I can start with a question not directly related to your work, what did you enjoy most about the book?"

TAUBENBERGER: "I liked the character development; I think the main characters were very well developed and 'real' appearing. I also liked the suspenseful aspect of the plot."

CRANE: "One of the plot elements in the book centers around the fictional recovery by a biotech company of a live sample of the 1918 Spanish Flu virus. Supposedly, in the story, the virus is isolated from a cadaver frozen for the last eighty years on a glacier in Alaska. In real life, a Canadian-led team has proposed visiting a cemetery located amid the permafrost of Longyearbyen, Norway, with a similar goal in mind. You have said elsewhere that, because flu-virus is so delicate, the chances are rather slim that the 1918 strain could exist in a type of frozen slumber for that length of time. As a cellular pathologist, can you foresee any conditions under which an intact sample of the Spanish flu virus might survive to the present day?"

TAUBENBERGER: "I see no chance for the virus to survive under any natural conditions. Permafrost fluctuates between about -10 C and +4 C, which is a temperature zone ideally suited to kill influenza viruses. [To have survived the last eighty years intact] would have required a sample to be frozen continuously in liquid nitrogen, or be stored in a -70 C freezer. I do not think either things existed in 1918. So in real life, I see no way to get live virus. However, I have no problems with the recovery of live virus in the context of a work of fiction---just like I had no problems with the fictional making of dinosaurs from DNA bits in amber."

Not a mosquito, but a lizard
preserved in Dominican amber

CRANE: "That reminds me about where I got the idea for basing the virus's recovery on a frozen tissue sample. A piece in Scientific American�, entitled "Jurassic Virus?", mentioned that--at least up until 1993--people had tried in vain to recover genetic snippets of the 1918 strain from human remains. And by 1996 your group had managed just that, recovering the first fragments of the 1918 genome--"RNA bits"--from a lung-tissue sample stored for the last eighty years with the Institute's collection of vintage autopsy specimens.
     I understand that when your paper came out in March of 1997 it attracted a good deal of attention from both inside and outside the scientific community. Can you tell me something about why this should be so? If the level of public interest has been high, what do you think is the reason for our fascination with the Spanish Flu?"

TAUBENBERGER: "There has been (and continues to be) intense media interest in our work on the 1918 influenza virus. I guess there is something about the story which really resonates well with the press and the public's interest. One element is the fascination [arising from the possibility] that genetic analyses could be performed on eighty year-old autopsy tissues. The second aspect is that here is one of the greatest medical mysteries of all time: a lethal virus spreading around the world killing tens of millions of people in a few months. The third facet of interest lies in the possibility that a new influenza pandemic, and specifically one of 1918 magnitude, could happen again.
     I have often been asked what is the likelihood of a pandemic like 1918 coming again. The answer is still unknown. What is known is that influenza pandemics occur with surprising regularity, every 20-30 years. The chance of a new pandemic emerging is 100%. The last one was in 1968, so the chance of a new pandemic occuring in the near future is pretty good.
     We still do not know yet why 1918 was so lethal, but whatever the circumstances were that came together to make the strain such a killer, it could certainly happen again. The race is to find those features [of the virus] which will help us to predict when a newly emerging strain might share [a similar pathogenicity]. The scare over the H5N1 strain in Hong Kong stresses the urgency of this concern."

CRANE: "What was the breakthrough in the lab that allowed you to succeed in recovering 'ancient RNA' ? Could the technique be stretched to older material? For instance, the Egyptians did a wonderful job of preserving bodily remains. Any chance of seeing the remnants of RNA-virus infections there? Or is that an incredibly remote possibility?"

TAUBENBERGER: "Our lab spent a lot of time working out how to isolate RNA from 'suboptimal tissues' both formalin-fixed pathology material and poorly preserved necropsy tissue from dolphins in the 1987 and 1993 mass mortality events on the U.S. Atlantic and Gulf of Mexico coasts, respectively.
     Our interest in RNA isolation from fixed biopsy material was to develop useful clinical molecular pathology tests using tissues for which fresh samples were unavailable. We can now do this successfully and make diagnoses of lymphoma and leukemia, other tumors, and many viral infections, which has a huge impact on patient care.
     In regard to the dolphin research (funded by the EPA) we were able to sequence and characterize morbilliviruses in these tissues and help make the link between mass dolphin mortalities and infections by viruses closely related to canine distemper virus and human measles. [References to the Institute's dolphin morbillivirus work can be found below]
     As for the mummies, your guess that RNA extraction would be 'an incredibly remote possibility' seems like a very good answer. DNA is however possible and has been done by Svante Paabo and others."

CRANE: "You are due to publish a second paper shortly in which you give the complete base sequence for the 1918 gene coding for hemagglutinin--the protein used by influenza A virus to attach itself to the host cell. How difficult was the work? I can imagine trying to reconstruct a dropped vase and finding the job hardest where it came to locating the smallest pieces and recognizing where they fit in. Do similar problems turn up in your work, or are there enough big overlapping RNA segments to facilitate the reconstruction?"

TAUBENBERGER: "It has been a very difficult task. Ann Reid in the laboratory has been the only person doing the benchwork on the 1918 flu project. The hemagglutinin gene has 1778 bases and the largest fragment we can amplify is about 150 bases long. There are no larger overlapping fragments. We have to design our PCR reactions so that the sequences overlap to cover the primer sequences. You need to know the sequence of something to do PCR, but we don't know the sequence for the 1918 virus. Therefore we have to design 'degenerate' primers, containing more than one base at some positions, which allows it to amplify all H1 hemagglutinins, e.g., human, swine, bird. Therefore, for every PCR reaction, we may only get 35-100 bases of useful sequence after 1-2 weeks work.
     We were also slowed down by getting two new cases. Before proceeding any further, we decided to confirm that the gene sequence from the first case matched those of the two new cases. It is basically the same in all three, with only very minor changes."

CRANE: "Now that the coding for the hemagglutinin gene is known, it should be possible to reengineer a strain of influenza virus which expresses the 1918 surface protein. By adding genes one by one, it is also possible to imagine recreating the original strain by the time the sequencing effort is completed. Can you foresee the day when that might literally be accomplished?
     Science derives a large measure of its power through the sharing of data and open debate. But in a case like this--where a killer virus is involved--it would be hard to imagine you didn't have some reservations about publishing all your work. What are your thoughts on this issue, speaking either as a scientist or from a personal point of view (if they aren't in complete accord with each other)?"

TAUBENBERGER: "It is certainly possible to make recombinant influenza viruses containing partial or complete 1918 sequences. While it does concern us somewhat, we feel that the positive aspects of the work--finding out where the 1918 strain came from, and why it was so lethal--far outweigh any concerns. Now that the complete hemagglutinin sequence is known, a protective vaccine can (and will) be made. Making such recombinant viruses may be necessary in controlled lab settings to understand the behaviour of the 1918 virus."

CRANE: "Where do you see your work going?"

TAUBENBERGER: "We are going to generate the complete gene sequences of several more 1918 flu genes, including neuraminidase, nucleoprotein, matrix and nonstructural genes. This will help us get a handle on the two big issues: Where did it come from? and, Why was it lethal?
     Did it come directly from birds to humans as in Hong Kong in 1997, or did it go through an intermediary like swine? The hemagglutinin sequence already begins to shed light on these questions, but each gene needs to be examined to see the complete picture emerge. Unfortunately, this will take a long time, several years at least."

End of interview

Leonard Crane would like to express his appreciation to Dr. Jeffery Taubenberger for taking the time to participate in this interview.

Follow up:

When this interview was first published in 1998, work had already begun on the reconstruction of the 1918 Spanish Flu virus by Dr. Jeffery Taubenberger and his team at the Armed Forces Institute of Pathology. In October of 2005 it was reported that the virus had been finally reconstructed after 10 years of work. When I interviewed him I had forseen the day when this would happen. You can read about the announcement in the PBS article entitled 1918 Spanish Flu Offers Clues About Pandemic Viruses.