LONDON - The prospect of a vaccine against influenza which would confer life-long protection against all strains of the virus is an exciting one. Now a new strategy, focusing on a tiny surface antigen of the influenza virus, suggests that such a vaccine may one day be possible.

The variability of the two main proteins of the influenza virus - neuraminidase and hemagglutinin - is notorious. Regular small changes due to point mutations, known as antigenic drift, account for the virus's ability to outwit the immune systems of people who have had influenza just one or two years previously.

Larger occasional changes, involving the substitution of a human influenza gene with one from an animal influenza virus - so-called antigenic shift - account for the influenza pandemics. When this happens, very few people have any immunity to the new viral strain, and the virus spreads rapidly around the world. In the Spanish flu pandemic from 1918 to 1920, for example, about 20 million deaths from the disease occurred in Europe alone.

For vaccine designers, it is like constantly trying to hit a moving target. These difficulties led a group of Belgian researchers to focus on a different viral protein - that known as M2. This small protein spans the viral membrane. It is present in only small amounts on the viral surface, but much more abundant on the surface of viral-infected cells.

Unlike neuraminidase and hemagglutinin, however, the gene encoding the extracellular portion of M2, a peptide known as M2e, is highly conserved. With only two exceptions, the amino acid sequence of this peptide has remained the same in all the human influenza A strains isolated and studied since 1933.

Walter Fiers, professor emeritus of molecular biology at the University of Ghent, told Bioworld International, "M2e is a small domain but it has the advantage that the amino acid sequence is virtually constant, independent of shift or drift. What we wanted to do was to present this small domain in such a way that its efficacy in stimulating the immune system would be much enhanced. We believe that we have succeeded in that."

Fiers, working with Sabine Neirynck and colleagues at the University of Ghent and the Flanders Inter univer sity Institute for Biotechnology in Ghent, Belgium, reports the team's results in the October 1999 issue of Nature Medicine in a paper titled "A universal influenza A vaccine based on the extracellular domain of the M2 protein."

The group's strategy was to take the gene for M2e and link it to the gene for the hepatitis B core protein. They then expressed the fused genes in Escherichia coli, under the control of a suitable promoter. What the bacteria made was virus-like particles of hepatitis B core protein, with the particles'9 surfaces covered with M2e.

The particles were purified and used to vaccinate mice. Initial experiments showed that mice given a dose of 10 micrograms of antigen with adjuvant, as an intraperitoneal injection, were protected against a dose of influenza virus that would normally be lethal. Fourteen out of 14 vaccinated mice survived a lethal challenge, whereas all mice in the control group died.

Tests on the lungs of vaccinated mice which had been challenged with influenza virus showed that there was a 98 percent reduction in the amount of virus present in the lungs. Fiers and his colleagues write, "These results demonstrate that vaccination . . . considerably enhances virus clearing from the lungs."

Further experiments showed that intranasal administration of particles (without adjuvant) was as effective in protecting the mice as intraperitoneal injection, and that the protective immunity could be transferred by serum, and was therefore due to antibody production.

In a comment on the paper published in the same issue of Nature Medicine, Edwin Kilbourne, of the Department of Microbiology and Immunology at New York Medical College in Valhalla, New York, said, "Encouraging as these results are, it may be premature to herald this vaccine as 'universal.'" His article, titled "What are the prospects for a universal influenza vaccine?" points out that the study used only two influenza viruses, both with the same M2 protein.

But Fiers said the results represent an important step toward a universal vaccine, adding that "the target of the protective immunity is the external domain of the essential M2 protein, which is virtually invariant in all human influenza A strains isolated so far."

He told Bioworld International, "If the present results in mice can be extrapolated to humans - and I can't see a good reason why this should not be possible, but it has to be shown - then it could be very significant. It could give us a vaccine which gives life long protection against all strains of influenza, just like those which protect against polio, say, or hepatitis B."

Although Fiers himself is now retired, he expects that future work at Ghent will examine ways of further optimizing the immune response triggered by the vaccine, perhaps by including influenza-specific T-helper or cytotoxic T lymphocyte epitopes into the recombined M2/hepatitis protein.