Newly discovered flu molecules shared by most strains of the virus could help scientists develop a "universal vaccine", it was claimed today.
Researchers identified the molecules after after subjecting healthy volunteers to flu infections.
They found that participants' immune systems targeted a specific range of peptides, or protein building blocks, within the internal structure of the flu virus.
Harnessing the immune system's response to the peptides could produce an all-encompassing multi-strain vaccine, the scientists believe.
Current flu vaccines produce an antibody response to surface molecules which alter rapidly to keep one step ahead of the immune system.
But the internal peptides only change very slowly and do not vary between strains. They also trigger a response from T-cells - white blood cell elements of the immune system - rather than antibodies.
A T-cell vaccine aimed at the molecules has the potential to provide long-term immunity against all major flu strains, including seasonal, avian (bird) and swine viruses.
Past flu pandemics have caused disaster on a global scale. The 1918 Spanish Flu pandemic killed up to 40 million people, more than the number who died in the First World War.
Britain's last pandemic outbreak of swine flu in 2009 claimed 457 lives. Experts believe more severe pandemics may occur in the future, possibly as a result of a mutated avian (bird) flu strain.
Study leader Dr Tom Wilkinson, from the University of Southampton, said: "Influenza is a virus that we know has a global impact, and the threat of further pandemics is a real one.
"Most influenza vaccines only protect us against known influenza strains by creating antibodies in the blood but the influenza virus has the ability to rapidly change itself and new strains can emerge which rapidly spread across the globe by escaping this immunity.
"We have found that there is an important role for T-cells that recognise the flu virus, which if harnessed could protect against most or even all strains of seasonal and pandemic flu.
"Through this discovery we hope to improve vaccines for future strains of influenza, and potentially protect against the next pandemic. However there is more to do to translate these findings into new approaches to treatment."
The research is published in the latest on-line edition of the journal Nature Medicine.
Co-author Professor Sir Andrew McMichael, director of the Medical Research Council Weatherall Institute of Molecular Medicine at Oxford University, said: "Current flu vaccines are very good at producing antibodies against flu, but not so good at generating a lasting immunity involving T-cells.
"The big question is: if we had a pandemic involving a much more severe virus than the swine flu we saw, what would we do in the six months it takes to develop an effective vaccine? This study suggests that vaccines stimulating a T-cell response might be an option, but there remains a lot to do to be certain of this approach."