Nearly half a century later, Dr Jeffery Taubenberger, chief of the division of molecular biology at the Armed Forces Institute of Pathology, was casting around for a new research project. The AFIP, located on the campus of the Walter Reed Army Medical Center in Washington, had been established by a Civil War general in 1862 as an army pathology museum designed to combat ‘diseases of the battlefield’. With its vast library of tissue samples and expertise at analysing tissues for the diagnosis of disease, the AFIP was a valuable resource for researchers and clinicians alike, regarded as ‘a stalwart of the international biomedical community’.10 Every year, the AFIP received at least 50,000 requests for second opinions on difficult cases from external pathologists, and its employees, including experts in many areas of human and animal pathology, ‘made major or minor changes to roughly half of the cases they acted on’.11
Although the director still reported to the Surgeon General of the Army and not the commander at Walter Reed, the institute had expanded its research remit. In 1991, Tim O’Leary was appointed chair of the department of cellular pathology and set out to create a ‘molecular diagnostic pathology component, with the idea that molecular biology tools would be useful as an adjunct in anatomical surgical pathology’.12
As a result, Jeffery Taubenberger and his team were engaged on research projects into the genetic structure of breast cancer and the recovery of the RNA virus from the decomposed flesh of dead dolphins, to determine whether the dolphins had died of a measles-type virus or a condition known as ‘red tide’, a harmful algal bloom toxic to sea life. Taubenberger’s team, consisting of doctors Amy Krafft and Thomas Fanning and microbiologist Ann Reid, were given the challenge of resurrecting the virus from several decomposed tissue samples using a technique known as PCR or Polymerase Chain Reaction.
When the AFIP was threatened with funding cuts in 1995, Taubenberger and his colleagues responded by looking at how they could apply PCR to the immense warehouse of tissue samples at the institute.
‘It was just kind of a hobby project idea,’ Taubenberger admitted. ‘I had a brainstorming session with Tim O’Leary, and the idea that we came up with was to go after the 1918 flu … the 1918 flu would be something that would be enormously useful, potentially of practical importance.’13
Taubenberger had learned briefly about the Spanish flu pandemic of 1918 in medical school, but knew little about it. After reading Alfred Crosby’s America’s Forgotten Pandemic, Taubenberger became intrigued by the sheer scale of the pandemic and the way it had faded so quickly from the ‘cultural memory’.14 Working for an army research laboratory, Taubenberger was particularly astonished by the toll taken on the US army.
The US entered World War I very late compared to the European combatant countries, so there were far fewer US military casualties – about 100,000 in total in World War I. But of those, over 40,000 died of influenza. So 40% of people who were young, healthy, well-fed, 18 to 25-year-old strapping American GIs dropped dead of flu in 1918. Which is an absolutely unbelievable number, when you think about that!15
Resolving to find the remains of the Spanish flu virus, Taubenberger ordered up tissue samples from seventy-seven soldiers who had died in the pandemic. Taubenberger’s team searched for samples of victims who had succumbed to the initial viral infection and not the subsequent bacterial pneumonia. Seven samples seemed promising.
It was an exciting moment when the cases showed up on his desk. ‘There they were!’ he said. ‘Having not been touched in 80 years!’16
These samples consisted of tiny scraps of lung tissue, preserved in formaldehyde embedded in pieces of paraffin wax. Taubenberger and his team then attempted to ‘develop techniques to tease out fragments of the virus’s genome from these tissues’.17
After seventy negative results, the researchers almost abandoned the project, but it proved so fascinating that they were reluctant to stop. ‘And the more we read about this virus and the outbreak and the devastating impact it had, the more committed we became to get the project to work.’18
By this stage, research into the causes of Spanish flu had become something of an obsession for Taubenberger. ‘It just got impossibly too much,’ he admitted. ‘Breast cancer was the first to go, then dolphin viruses and T-cell development. And flu just kind of took over my life!’19
Research into the origins of the Spanish Lady was also lengthy and time consuming. ‘This kind of “viral archaeology” is extraordinarily painful! And very slow,’ Taubenberger commented. ‘It’s not like you can just go to the freezer and pull out the virus. We have to find autopsy cases.’20
The breakthrough came a year later, in July 1996. Dr Amy Krafft began examining a sample of preserved lung tissue from the body of Private Roscoe Vaughn. Private Vaughn had been just twenty-one years old when he fell sick on 19 September 1918, at Fort Jackson, South Carolina. A week later, at 6.30 a.m. on 26 September, Private Jackson died of pneumonia. At 2 p.m. that same day an autopsy was performed on Private Vaughn’s body and a pathological specimen removed from his lung.21 Now, eighty years after his death, Private Vaughn was to play his part in helping to solve the mystery of the Spanish Lady.
Vaughn’s contribution to medical science lay in the unusual manner in which his body had responded to Spanish flu. Taubenberger was struck by the way his case looked under the microscope.
What was recognized clinically is that he had developed a pneumonia of his left lung [and the right lung was seemingly normal]. Now it’s quite common to have pneumonia of just one lung, not necessarily both. And at the time of autopsy that was confirmed – he had massive bacterial pneumonia of his left lung that was fatal, and his right lung was almost completely normal.22
This went unnoticed at the time of the autopsy, ‘but what was there if you looked at the sections quite carefully were little tiny areas of very acute inflammation around the terminal bronchioles in that lung that were characteristic of very early phases of the influenza viral replication’.23
Private Vaughn was also unusual because he demonstrated an ‘asynchrony’ in the way that the disease developed,
in that he got influenza infection, and then had a bacterial pneumonia in his left lung which overtook the viral infection and killed him, but the influenza virus infection of his right lung was somehow delayed by several days. So when he died it left a snapshot of the very earliest stages of the virus infecting the lung. It was a very subtle change, and it took me a while to look at enough autopsies to get a sense of that. Influenza does not have characteristic changes that allow you to be confident, to just look under the microscope and say, ‘This is definitely influenza.’ You can sort of suspect it, but you can’t be certain. But there was something about that case that just struck me as an excellent example, and so once I’d identified it, we extracted RNA, did our test and, boom! We found influenza virus RNA!24
Given that the quality of genetic material recovered was, in Taubenberger’s words, ‘horrible!’ the team had the daunting task of attempting to carry out ‘large-scale sequencing of the virus to try to work out what it was’.25
Concerned that there was not enough material for testing, Taubenberger put aside the material from Private Vaughn and looked for other cases. Another specimen was located, which had come from another person who had died on the same day, but in a different camp. This too was positive, so the team now had two cases.26
Taubenberger and his colleagues published their initial findings, ‘just little tiny fragments of sequence of the virus from our pathology material from the AFIP’,27 in Science magazine in 1997. It was at this point that Johan Hultin, now aged seventy-three, read the article and wrote to Taubenberger, explaining his own research. When Taubenberger requested the opportunity to look at Hultin’s old material from Alaska he discovered, to his horror, that it had been destroyed, just a few years earlier.
Hultin proposed that they return to Alaska to exhume more bodies and obtain further frozen material, upon which they could perform molecular analys
is at Taubenberger’s AFIP laboratory, an activity that would not have been possible in 1951. The structure of DNA had not even been determined until 1953. There was no stopping Hultin. Despite being seventy-three years of age, he set off for Brevig Mission, equipped with little more than a camera, a sleeping bag and two bags of tools. Hultin funded the entire expedition himself, but it was a decidedly low-tech affair and Hultin slept at night on the floor of the local school. Taubenberger did not accompany him, and perhaps this was advantageous, as, given the sensitive nature of the work, many people in the community remembered Hultin and were content to allow him to excavate. ‘They had been children and they were now the elders in this community,’ said Taubenberger, ‘remembering him 45 years later. He got permission to do an exhumation and he sent us material.’28
Brevig Mission’s seventy-two influenza victims had been buried in a mass grave in 1918, marked with two large wooden crosses. With the help of four young men, Hultin began to dig until they had excavated a trench twenty-seven feet long, six feet wide and seven feet deep. At first it seemed as if, once again, Hultin was to be disappointed. He discovered skeletons, but no remains containing soft tissue. But then they uncovered the body of an obese woman, whose fat had preserved her organs from the depredations of permafrost.
‘I sat on a pail and looked at this woman. She was in a state of good preservation. Her lungs were good. I knew that this is where the virus would be.’29
Hultin named the woman ‘Lucy’, after the famous prehistoric skeleton recovered in Ethiopia in 1974. Taking samples of Lucy’s organs, Hultin packed them in preservative. To prevent the samples from being lost, Hultin sent four identical sets over four days, courtesy of UPS, Fed Ex and the US Postal Service. He closed the grave and planted two new crosses there, which he had made himself in the woodwork department of the local school.
Hultin had made a significant discovery. Although three tissue samples proved negative, Ann Reid identified traces of the influenza virus in ‘Lucy’s’ samples. They proved identical to the RNA from Private Roscoe Vaughn. Soon afterwards, a third match was made from a sample from Private James Downs, who had died of influenza at Camp Upton, New York, on 26 September 1918, aged thirty. It was ‘Lucy’, however, who provided the missing link.30
‘Using the frozen material it became possible to sequence the entire genome of the virus,’ said Taubenberger.
We ended up sequencing the hemagglutinin gene, the sort of main gene of the virus, from all three of the cases. And what we found, amazingly, was that they were basically identical, one to the other – that out of 1700 bases in this gene, these three cases differed from each other by only one nucleotide. So we knew that this was really the pandemic virus; there was no question that this was the virus. And using the frozen material it became possible to sequence the entire genome of the virus.31
Hultin’s earlier research was vindicated: he and Taubenberger’s team had found the Spanish flu virus.
‘It took an enormous amount of effort on our part, from 1997 through early 2005, to fully sequence the genome of the virus,’ said Taubenberger. ‘One of the things we concluded from this study was that this was a sort of bird flu virus that adapted somehow to humans.’32
How that pandemic virus was transferred from animals to people is another question which is still hotly debated. Taubenberger, in common with other virologists, is at the forefront of trying to find the answer to this dark riddle, equipped with a technique that enables scientists to create influenza viruses from cloned genes. A number of scientists arrived at this technique, known as ‘reverse genetics’, simultaneously. ‘Drs. George Brownlee at Oxford, Peter Palese at Mount Sinai, and Yoshi Kawaoka in Wisconsin kind of all independently worked on this,’33 Taubenberger explained.
Having sequenced the virus, ‘through the wizardry of modern molecular biology’,34 scientists then tested it on animals in ‘high containment’ laboratories at the Center for Disease Control in Atlanta and at a laboratory in Winnipeg, Canada, where the 1918 virus has been put into macaques.
As to what made the Spanish flu pandemic so deadly, and why the virus killed so many healthy young people, Taubenberger subscribed to the theory that the virus provoked an auto-immune response known as a cytokine storm. Ironically, the healthier the patient, the more likely they were to die. The 1918 H5N1 produced a marked inflammatory response, causing secondary damage to the patients’ lungs. ‘It’s not the virus that kills you but your own body’s immune response,’ explained Taubenberger.35
Mercifully, the twentieth century has not seen another pandemic on the scale of 1918’s Spanish flu. In 1957, there was an outbreak of Asian flu (H2N2) and in 1968 of Hong Kong flu (H3N2). There was a ‘pseudopandemic’ in 1947 with low death rates, an epidemic in 1977 that was a pandemic in children, and an epidemic of swine influenza in 1976 which was feared to have pandemic potential.36 But the 1918 pandemic appeared to have been a unique event, a combination of circumstances which, it was hoped, would never occur again. As a result, by 1997 research into the origins of Spanish flu might have been regarded as a somewhat recherché project, an academic quest with little relevance to contemporary life. Research funding was in jeopardy, with the finance director’s blue pencil hovering over the APIT budget. But then, just as Taubenberger and his team were publishing their first research findings in March 1997, a three-year-old child in Hong Kong got infected with an H5N1 bird flu virus and died.
CHAPTER TWENTY-TWO
THE HONG KONG CONNECTION
ON 9 MAY 1997, a little boy fell sick in Hong Kong. Normally lively and robust, three-year-old Lam Hoi-ka was struck down suddenly with a fever and sore throat. Lam’s anxious parents called their doctor only to be told that Lam was experiencing a routine childhood malady and would recover within a day or two. Lam’s symptoms suggested an upper respiratory infection, common among children the world over and the kind of illness a busy doctor sees dozens of times a day.
But, five days later, Lam had not recovered and so his parents took him to the local community hospital.1 Staff there could not identify the cause of his symptoms but they were sufficiently concerned to admit Lam to the Queen Elizabeth Hospital in Kowloon, where the doctors were once again unable to make a diagnosis. But something clearly was wrong: Lam was going downhill rapidly. The little boy could not breathe without a ventilator, and the best guess was that he was suffering from viral pneumonia. As if this was not bad enough, Lam had also developed Reye’s syndrome, a rare disease which generally afflicts children and teenagers and can be fatal. Reye’s, which often succeeds viral infections such as influenza or chicken pox, causes fluid on the brain, which puts pressure on the nerves that control breathing and heart rate. Once this happens, the patient dies.2 Lam was dosed with antibiotics to treat his pneumonia, but then developed ‘disseminated intravascular coagulopathy’, in which the blood clots like curdled milk. Normal clotting is disrupted, leading to severe bleeding from several sites.3 Lam suffered massive organ failure and died a week after he had been admitted. Lam’s devastated parents and shocked medical staff were left wondering how on earth, in the final decade of the twentieth century, a healthy little boy could fall sick and die so quickly.
On 20 May 1997, the day before Lam Hoi-ka died, doctors took a specimen of throat wash from his windpipe for analysis. This specimen went to Hong Kong’s Department of Health for testing, as a matter of routine. Laboratory staff investigated the specimen and concluded three days later that Lam had died of influenza. However, despite extensive testing, chief virologist Dr Wilina Lim could not determine the type of influenza that had killed the little boy. Tests had eliminated H3N2, the descendant of Hong Kong’s 1968 epidemic. Nor was the H1N1 virus of an outbreak in 1977 to blame.4 Unperturbed, Lim sent the specimen to the World Health Organization’s Collaborating Centres devoted to research into deadly diseases, based in London, Tokyo, Melbourne and at the Center for Disease Control in Atlanta, Georgia. Like international terrorism, pandemic disease is a constant hig
h-level threat, and these centres are the early warning stations for pandemics, keeping a lookout for newly evolving strains of deadly viruses, including influenza, SARS and Ebola. Thousands of samples are sent every year for testing. Lim also sent a sample to eminent virologist Jan de Jong at the Dutch National Institute of Public Health near Utrecht.
On Friday 8 August, de Jong rang Lim telling her that he was flying straight to Hong Kong, but not saying why. When Lim picked him up from the airport, de Jong explained the reason for his sudden visit.
‘Do you have any idea what virus you sent me?’ he asked her.
Lim replied that she thought it was an H3, but one that had evolved so much that she could not identify it through laboratory tests.
‘No,’ de Jong said. ‘It was an H5.’5
As expert virologists, they both knew what this meant. H5 was bird flu, but bird flu that had just killed another human being. Was Hong Kong on the verge of a pandemic?
Meanwhile, at the Center for Disease Control in Atlanta, Nancy Cox, chief of the influenza division, had just returned from vacation and resumed analysis of the thousands of specimens sent in from all over the world. As far as Cox was concerned, the Hong Kong sample was just another specimen, waiting its turn. A month passed before Cox’s team examined it. When Cox saw the test results, she was horrified. Just like de Jong, Cox confirmed that Lam Hoi-ka had died of bird flu, in Asia, long considered to be the epicentre of influenza by scientists.
Cox’s first duty was to protect her staff. The research operation was moved to a biosafety level three-plus containment facility and their existing protective equipment was increased so that they worked in heavy layers of hoods and masks, like the modern equivalent of Jacobean plague doctors.6 The similarities did not end there, with Cox fearing that the world was poised on the verge of a deadly plague.
Pandemic 1918 Page 24