Mammals - including horses, pigs, dogs and people - provide a dangerous melting pot for the influenza virus. And scientists are worried. Neil Clarkson reports.
Birds form the key global reservoir for the flu virus - and migratory birds can carry the disease around the globe.
There are many viruses that are specific to species, but influenza virus has shown a worrying trait: an ability to jump the species barrier into mammals.
Current thinking suggests that all influenza A strains that affect mammals originated from the wild aquatic bird population.
New strains jumping the species barrier worry disease experts, as humans are likely to have limited or no resistance to these new infections. The recent outbreak of equine influenza in Australia shows just how explosively influenza can spread in a population never before exposed to it.
Experience shows the infection rate among a naive population of horses such as that in New Zealand would be 60-90%, with a potential fatality rate of 1-20%, with foals and weaker animals most susceptible.
Humans have been badly affected by the flu virus. Ten to 20 human influenza pandemics are believed to have swept the globe in the last 250 years, the worst being the so-called Spanish flu of 1918 to 1919 that caused more than 20 million deaths and affected some 200 million people.
Research has shown that the Spanish flu virus derived directly from a bird strain.
The potential for serious loss of life in a new worldwide flu pandemic is high. For that reason, major efforts have been made to limit the spread and effects of avian influenza, particularly the H5N1 strain which has already shown an ability to infect people and cats.
But flu is not just a human problem - or a bird problem for that matter.
Flu viruses have mutated to affect horses, pigs and cats. Around 1999, an equine strain made the jump into dogs - the first time that horse flu was shown to have jumped the species barrier.
The fact that birds form the key global reservoir for the flu virus presents obvious problems. Migratory birds can carry the disease around the globe.
While New Zealand and other countries have yet to be exposed to bird flu, the possibility of migratory birds carrying it to our shores cannot be ruled out. Godwits, for example, are known to migrate north from New Zealand through areas where bird flu is prevalent. However, their return southern migration takes them further east through the Pacific, where the disease is not yet a problem.
Furthermore, evidence suggests birds moving longitudinally (north-south) play a bigger role in the continuing process of avian virus evolution than those moving latitudinally (east-west).
In horses, two major sub-types of influenza A virus are known to circulate. The first (H7N7) was isolated in Czechoslovakia in 1956, the second (H3N8) in Miami in 1963.
The H7N7 sub-type is believed to be extinct or present only at very low levels.
The H3N8 sub-type is almost certainly a mutation of a bird flu virus, and changes in the virus have seen the development of two lineages - American and European. The H3N8 sub-type was behind a severe 1989 outbreak in China, which affected 80% of horses exposed to it and killed 20%. It went on to cause outbreaks in Trinidad (1979), Argentina (1985), South Africa (1986), and Jamaica (1989). H3N8 was behind the recent Australian outbreak.
Migratory birds have the ability to spread the influenza virus far and wide.
Dalva Mancini and a team of researchers in Brazil have looked for evidence pointing to the transmission of influenza between horses and people. Their work revealed high levels of antibodies in horses to two human influenza A strains - H1N1 and H3N2.
The researchers had taken fluids and mouth and nasal swabs from 46 horses at studs and racing stables in Sao Paulo and Rio de Janeiro, in Brazil.
About 78% of the horses carried antibodies to two strains of equine flu. Remarkably, they tested positive in even higher numbers to human influenza A virus - 80.43% for H1N1 and 93.47% for H3N2. Similar numbers had antibodies to type B influenza.
The results meant that the immune systems of the horses recognised and generated antibodies against the human flu viruses, but did not become sick.
"A high incidence of human influenza virus infection was observed among the horses evaluated," the researchers said. "These results demonstrated that horses are reservoirs of the circulating influenza viruses of both the equine-specific and the non-specific strains."
"These interspecies transmissions reinforce the potential mechanisms for introduction of genes of the avian influenza viruses in reservoirs that can reach man.
"The response to the type B strain in the horse sera is particularly noteworthy, since this strain is considered to be restricted to humans," they said in their findings.
The researchers continued: "It may be concluded ... the high incidence of influenza viruses among animals and humans surely represents a problem to both public health and animal protection all over the world.
"The occurrence of interspecies transmission is of concern since this allows the maintenance of the influenza virus in different reservoirs."
The evidence, they said, led to the conclusion that interspecies transmission probably occurs between humans and horses.
G.A. Landolt, in a presentation to the American Association of Equine Practitioners in 2006, reviewed the transmission of influenza A viruses between species and its implications for horses.
He described equine flu as a disease that generated serious health and economic problems, adding that vaccination was the cornerstone of influenza prevention.
Vaccine failure, he said, could primarily be blamed on the sustained genetic evolution of the virus, called antigenic drift, as well as the occasional introduction of gene segments or entire viruses from other species, called antigenic shift. These factors, he said, accounted for the diversity of influenza viruses.
"Although the barrier to transmission of avian viruses to horses is relatively strong, it is anything but complete," he said.
A severe epidemic in horses in northeast China in 1989, he said, served as an example of such spread. Genetic evidence pointed to this virus having spread directly to horses from the bird reservoir.
It is possible that equine influenza could appear at any time - and poor biosecurity will not necessarily be to blame.
Further research has since shown this virus had been circulating in several states as early as 1999.
"The transmission of influenza from horses to dogs is especially interesting in light of the fact that dogs were not commonly regarded as hosts for influenza A viruses.
"The viral and host factors that determine the species range of influenza viruses and the mechanisms by which host barriers are overcome are not completely understood. Regardless of the mechanisms involved, the equine-lineage H3N8 viruses have since become established in the dog population in the United States," he said.
Although horse strains have shown relatively little genetic diversion, the new canine variety has shown high rates of evolution.
"Although it is unclear if the canine H3N8 viruses are still capable of infecting horses, they potentially represent a serious emerging disease threat to the American horse population.
"Although horses have often been regarded as isolated hosts for influenza, events of recent years clearly indicate that the species barrier for viruses jumping either to or from horses is not absolute.
"The realisation of this fact has importance for control, because commercially available vaccines may not provide protection against infection with viruses stemming from other species."
Dr Laurent Kaiser has also explored the question of influenza reservoirs and discussed the issues at an international symposium on influenza and other respiratory diseases in Memphis, Tennessee.
The main reservoir, he observed, was the wild aquatic bird population, which can transmit influenza viruses directly to species such as chickens and pigs.
Wild aquatic birds can spread influenza to pigs, and the two species should never be kept together in close quarters.
"Transmission of influenza viruses from birds to mammals has probably occurred for centuries. However, increased opportunities for transmission, larger chicken and pig populations, and overall growth of human populations are associated with a higher risk of interspecies reassortment.
"This situation is a possible start for a new pandemic. The timing of the next influenza pandemic cannot be predicted," he said, "but we know that it will occur eventually.
"As one noted expert said, 'The influenza clock is ticking, but we don't know what time it is'."
The US Centers for Disease Control suggest the next pandemic will result in 89,000 to 207,000 deaths, 314,000 to 734,000 hospitalisations, and between 18 million and 42 million outpatient visits in the US alone.
Influenza expert Dr Robert Webster, in a paper discussing the disease, said pandemic strains of human influenza emerged only rarely. However, interspecies transmission of influenza viruses may not be so rare. Up to 10% of persons with occupational exposure to pigs developed antibodies to swine influenza virus.
Most transfers of influenza viruses from pigs to humans are dead-end transfers; that is, they do not then spread efficiently from human to human.
Because all known influenza A subtypes exist in the aquatic bird reservoir, influenza is not an eradicable disease; prevention and control are the only realistic goals.
"The influenza virus continues to evolve, and new [strains] emerge constantly, giving rise to yearly epidemics.
"In addition, strains to which most humans have no immunity appear suddenly, and the resulting pandemics vary from serious to catastrophic."
Influenza viruses needed to be monitored not only in people, but in pigs and aquatic birds, he said.
"Moreover, if pigs are the mixing vessel for influenza viruses, surveillance in this population may provide an early-warning system for humans."
He said studies have led to the hypothesis that all mammalian influenza viruses derive from the avian influenza reservoir. Support comes from analyses showing that avian influenza viruses have evolved into five host-specific lineages: ancient equine - which has not been isolated in over 15 years - recent equine; gull; swine; and human.
There are, he said, few signs that avian flu viruses have continued to evolve in the past 60 years. This suggests the bird flus are approaching or have reached optimum, whereby changes provide no selective advantage.
"It also means that the source of genes for pandemic influenza viruses exists phenotypically unchanged in the aquatic bird reservoir.
"... the ancestral viruses that caused the Spanish flu in 1918 ... are still circulating in wild birds, with few or no mutational changes."
"Since the first human influenza virus was isolated in 1933, new subtypes of human type A influenza viruses have occurred: H2N2 (Asian influenza) replaced H1N1 in 1957, Hong Kong (H3N2) virus appeared in 1968, and H1N1 virus reappeared in 1977.
"Each of these new subtypes first appeared in China, and anecdotal records suggest that previous epidemics also had their origin in China."
Avian influenza viruses in wild aquatic birds are spread by faecal-oral transmission through the water supply, with initial transmission to mammals, including pigs and horses, probably also occurring through faecal contamination of water.
Domestic fowl should not be allowed to mix with wild waterfowl because of the risk of transmission of viruses.
People should be discouraged from living in close proximity to poultry and pigs - a factor believed to play a part in the generation of new strains in Asia.
The lessons from influenza's viral melting pot is clear. The virus is a moving target.
The handful of countries free of equine influenza should not think that the matter ends with good biosecurity.
There is so much more to consider: bird-flu outbreaks, the migratory patterns of birds, evidence that mammals are acting as reservoirs for flu viruses, and influenza's ongoing demonstration that it can jump the species barrier.
Today, only New Zealand and Iceland are considered free of the virus, with Australia likely to officially rejoin those ranks late this year.
However, it is possible that equine influenza could appear at any time - and poor biosecurity will not necessarily be to blame.
Nature's tiny virus may yet prove resourceful than the best biosecurity controls.