Science for Health
Rick Carver and John Skehel
A group of viruses called paramyxoviruses have become the scourge of numerous animal populations and the bane of conservationists. In the recent press, Nipah virus, named for one of the towns in North East Malaysia where the disease first occurred, has devastated pig farms in Malaysia and caused the deaths of over a hundred farm and abbatoir workers from the Chinese border in the North, to Singapore. Over 900,000 pigs have been slaughtered in attempts to control the spread of disease. Five years ago in Australia, a similar virus called Hendra virus, caused pneumonia in race horses in Hendra, a suburb of Brisbane. A trainer and thirteen out of twenty infected horses died. Both Nipah and Hendra are very like distemper viruses that were responsible for the death of over twenty thousand Harbour and Grey seals in the Caspian, Baltic and North Seas in 1988, and for the decline of porpoise and dolphin populations in the Mediterranean and European Atlantic since the early 1990s. In the Serengeti in East Africa, distemper virus also caused the death of about one third of the closely monitored lion population in 1994, and numerous but unknown numbers of hyenas, foxes and leopards also died. Unlike the Nipah and Hendra viruses, which seem to have transferred to domesticated pigs and horses from infected fruit bats, transfer of distemper in Africa was in the reverse direction, from infected domesticated dogs to the wild animals of the neighbouring National Park.
Before 1930, distemper was as devastating for young dogs throughout the world, infecting essentially all and claiming the lives of more than half of them. Infected animals have high temperatures and may vomit and have diarrhoea. There is a discharge from the nose and eyes and the disease is spread when virus in droplets of discharge is inhaled. The animals cough and may have pneumonia. About half of them have fits and tremble uncontrollably, even when they are asleep. Sometimes their hind legs are paralysed. If they survive, distemper can also cause long-term problems which include damage to tooth enamel, marked thickening of the nose and footpads, known as "hardpad" which is another name for the disease, and "old dog encephalitis" a sort of canine senile dementia.
The reason that so few veterinarians and dog owners have any experience of this dreadful disease is that since the 1930s, vaccination has become widely available and consequently the disease is now rarely seen in the United Kingdom. The development of this vaccine in the late 1920s was the first research objective at Rhodes Farm, an outstation of the National Institute for Medical Research in Mill Hill, and the site of its laboratories since they were moved from Hampstead fifty years ago.
In an enormously successful five year collaboration financed by donations from dog owners through an appeal by the editor of The Field magazine, pathologist Patrick Laidlaw and veterinarian George Dunkin proved that a virus was responsible for distemper and succeeded in protecting dogs and ferrets from the disease by vaccination. Their vaccine was made from minced tissues of infected animals that were filtered to remove tissue fragments and bacteria, and then chemically treated to inactivate the distemper virus. From numerous carefully controlled experiments they could show that vaccine prepared with virus from infected dogs protected dogs better than ferrets; virus from ferrets protected ferrets better than dogs. Long-lasting immunity was achieved by infecting the animals with untreated virus two weeks after injection of vaccine. No symptoms of disease were seen but the complete immunity known to exist in the few animals that recovered from distemper was established. It’s interesting that the use of this procedure of vaccinating with inactivated virus vaccine and then with infectious virus is now recommended for vaccination against poliomyelitis; the first vaccination gives almost complete immunization and decreases the risk of disease which can occasionally be caused when infectious sugar-vaccine virus is used alone. By 1929 Laidlaw and Dunkin had developed their distemper vaccine and these procedures for its delivery to a stage when it could be handed on to Burroughs Wellcome and Co. for production of the first canine vaccine on a commercial scale.
The interest of medical scientists in distemper was based in part on similarities between the disease and such human diseases as measles and influenza, all of which first involve infection of the respiratory tract. In the year that the Institute was planned in 1918, the post-war world was wracked by a great influenza pandemic. Just as in the case of distemper, viruses or bacteria or both were suspected to be responsible for the disease but the nature of the infectious agent was not discovered for another fifteen years, and Rhodes farm was again the venue for the discovery. This time it occurred simply by chance. During one of the most severe winters for influenza in 1933, a number of ferrets being studied for immunity in distemper vaccine trials were seen to share all of the symptoms of coughing and sneezing that the influenza-infected researchers were suffering. But this was not a distemper vaccination failure. The ferrets had caught influenza from the scientists and once again became invaluable in virus research, this time in the isolation of human influenza virus, in the proof of its responsibility for the disease and in the first attempts to prepare influenza vaccines.
The wealth of information that we now have about viruses has accumulated mostly since the 1960s. We know about their shapes, their chemical components, and about the processes by which they are made. In fact a lot of the knowledge about growth and survival of the cells that make up all organisms has come from studies of viruses. This is mainly because compared with cells, viruses are smaller and simpler to investigate and yet they depend for their replication on similar processes. In the early1930s the fact that viruses are smaller than bacteria was the principal characteristic by which they were described. Early comparisons of the sizes of infectious agents showed viruses to be the smallest; smaller than bacteria, protozoa and fungi. This was discovered because they are not removed from infected body fluids such as nasal discharge, blood and spinal fluid, by filtration through porcelain filters. They vary from less than a millionth of an inch in diameter for viruses such as polio, to about three millionths of an inch for smallpox virus, compared with the sizes of bacteria which range from five to fifty millionths of an inch. In the early 1930s William Elford in the Institute brought these measurements to a stage of accuracy and reliability previously unattained, and it was largely due to his research that influenza virus was characterized. Elford became renowned for devising methods to form very fine films of cellulose that could be used as filters. Using mixtures of cellulose dissolved in different alcohols he found that films could be formed by evaporation, with precisely different pore sizes and these allowed detailed comparisons of the sizes of numerous infectious agents. Extracts of influenza-infected lung fluid were among the first to be examined and the infectious agent was found to be more than ten times smaller than the smallest known bacterium. The influenza virus was discovered.
These early studies at Rhodes farm were the beginnings of virology in the Institute which has always included research on one or other aspect of influenza. Among the most important early discoveries with the new viruses was the observation that they changed slightly from year to year and that the changes could be easily detected in laboratory tests using sera collected from infected ferrets. We now know that these changes can decrease the effectiveness of influenza vaccines and to this day the antibodies produced by ferrets infected with the most recent influenza viruses are used in tests that distinguish current viruses from those of previous years. These studies provide the basis each year for updating influenza vaccines.
To spot new viruses when they cause disease anywhere in the world a network of laboratories, with the Institute as the World Influenza Centre, was set up for the World Health Organization in 1947 by Christopher Andrewes, who with Patrick Laidlaw and Wilson Smith was one of the discoverers of the human influenza virus in 1933. From research in the last twenty years with American and French colleagues we now know in some detail the nature of the differences between viruses of different years. They mainly involve small changes in the shapes of proteins that project from the virus surface. Microscopy has shown that influenza viruses are spherical with diameters less than a millionth of an inch, covered with spiky protein projections like those on the surface of a horse chestnut conker shell. During infection the projections stick the viruses to cells of noses and throats and lungs and puncture the cells’ surfaces to release the virus genetic material into the cell so that it can be replicated for the next generation of viruses. When we recover from influenza or when we are vaccinated, our immune systems produce antibodies that protect us from being infected again by the same virus. They do this by binding to the virus spikes, preventing the viruses from sticking to cells. If, however, a virus with differently shaped spikes comes along then our antibodies are unable to bind to them and we have no protection. Hence the importance of having correctly shaped spikes in the vaccine and international surveillance to detect any new viruses so that the vaccine can be updated. In most years the small changes that occur in the projecting proteins only make them able to avoid recognition by the antibodies that some of us have produced, and then only some of us get infected. But from time to time quite unpredictably, an influenza is detected that can infect everyone. This happened in 1957 and again in 1968 causing the Asian and Hong Kong influenza pandemics. The new viruses responsible were named for the geographical locations where they were first detected. But where did the epidemics actually start ? And where did the viruses come from?
Increasingly it looks as if many of the new viruses come from the Far East, particularly from China. Retrospective studies of the dates of influenza outbreaks and their locations suggest that this was the case for the 1957 and 1968 pandemics, and probably also for the one in 1918, and year after year new viruses are identified first by the Chinese laboratories and subsequently are detected around the world. Population density, extremes of climate, and farming practices, particularly those on poultry and pig farms, have all been proposed as important and characteristically Chinese factors influencing geographical origin, but none of the explanations are as yet convincing.
The question of the origin of new human influenza viruses was taken up by Helio Pereira in the 1960s when he was Director of the World Influenza Centre through his special interest in animal and avian influenza. It had been appreciated since shortly after the great pandemic of 1918 that pigs could catch influenza with symptoms very similar to ours. Horses also were known to suffer from influenza pneumonia, occasionally with disastrous consequences for racing programmes. In the 1960s evidence began to accumulate for the widespread, mostly symptomless infection of wild birds, especially ducks and geese, and sometimes much more dramatic outbreaks in flocks of domesticated chickens and turkeys. Research had shown that the horse and bird viruses were quite different from those isolated from infected humans and the focus at Mill Hill was on mechanisms by which viruses might be transmitted between species. Special attention was paid to the possibility for interchange of genetic information between viruses from different species. This had been shown to occur very frequently when two different human influenza viruses were mixed together to infect cells in the laboratory. The genetic information of a virus like that of a human is present in a collection of genes, but a much smaller collection; in the case of an influenza virus ten genes compared with about fifty thousand in humans. It’s because of this comparative simplicity that viruses are unable to increase in number without infecting a cell to access its complex molecular machinery for such essentials as protein synthesis and energy production. The genes of most viruses are also like ours, linked together in long chains, but in a minority of viruses that includes influenza viruses the genes are unconnected separate molecules which are packaged together in complete sets in each spherical virus. When two influenza viruses, one from a bird and one from a human, infect one cell, as can be arranged in the laboratory, viruses with mixed sets of genes are produced. Among these mixed viruses Pereira and his colleague Bela Tumova were able to show that some had derived from the human virus the gene which codes for the protein that forms the virus surface projection, and had acquired all of their other nine genes from the avian virus. Should similarly mixed viruses with all the characteristics of a human influenza virus, but with the surface projection protein of an influenza virus from another species occur by chance in nature, they speculated, we would have no immunity to them. Some years later, when the genetic material of the 1968 Hong Kong pandemic virus was completely analyzed, its origin proved to have been almost exactly as they had predicted. Two genes of the Hong Kong virus, one of which coded for the surface projection, had come from an avian virus and the other eight were from the human virus that had circulated the year before in 1967. A new virus had been produced which was well adapted to replicate in humans but against which no-one was immune.
Not all cross-species infections by influenza viruses have occurred by this mechanism of gene exchange. In the late 1980s thousands of harbour seals died of pneumonia in the Western Atlantic, following infection with an avian influenza virus previously notorious as the virus that causes Fowl Plague. Around the same time in North Eastern China over ten thousand horses died of influenza following infection with a virus previously detected in wild ducks. In humans also in 1997,the size of an epidemic of influenza in Hong Kong in which six out of eighteen people infected with a chicken virus died, may well have been limited by the slaughter of all the chickens in the island’s farms and markets. And again in 1999 a number of less serious cases of influenza caused by transfer of a wild goose virus to humans were detected in South China and Hong Kong.
In all these incidents avian influenza viruses are acting just like the parainfluenza viruses distemper, Hendra and Nipah, as examples of so-called emerging infections that attract so much attention because of their great impact on the new populations that they attack. We have no protection against them. We cannot hope to eradicate them by vaccination as was achieved with smallpox and hopefully soon with poliomyelitis. In both of these cases the diseases are caused by viruses that only infect humans. We can only watch out for them and be prepared to restrict their spread and their most extreme consequences. In the year of the report of the BSE inquiry we must certainly expect the continuing story of infectious diseases to catalogue inter-relationships between infections of animals and of man. The early history of virus research at Mill Hill was certainly based on such expectations. From the very first studies on distemper to create a vaccine for dogs which is now important to protect endangered species, to the chance sneeze that infected a ferret and led to the discovery of the human influenza virus, there can rarely have been a more successful interaction between veterinary and human medical research.
This essay was published in the Mill Hill Essays 2000
© MRC National Institute for Medical Research
The Ridgeway, Mill Hill, London NW7 1AA